gretl-common 1.9.6-1build1 / usr / share / gretl / genrcli.hlp.pt

## Acessores
# $ahat
Output:     series

Tem que ser acedido após a estimação de um modelo de dados de painel com
efeitos fixos. Retorna uma série contendo as estimativas dos efeitos fixos
individuais (interceptores unitários).

# $aic
Output:     scalar

Retorna o Critério de Informação de Akaike para o último modelo
estimado, se disponível. Consultar the Gretl User's Guide para detalhes
sobre o cálculo.

# $bic
Output:     scalar

Retorna o Critério de Informação Bayesiano de Schwarz para o último
modelo estimado, se disponível. Consultar the Gretl User's Guide para
detalhes sobre o cálculo.

# $chisq
Output:     scalar

Retorna a estatística qui-quadrado global para o último modelo estimado,
se disponível.

# $coeff
Output:     matrix
Argument:   s (name of coefficient, optional)

Sem argumentos, $coeff retorna um vector coluna contendo os coeficientes
estimados para o último modelo. Com o argumento opcional do tipo texto
retorna um escalar, designadamente o parâmetro estimado com o nome s. See
also "$stderr", "$vcv".

Exemplo:

	  open bjg
	  arima 0 1 1 ; 0 1 1 ; lg 
	  b = $coeff               # obtém um vector
	  macoef = $coeff(theta_1) # obtém um escalar

Se o "modelo" em questão é realmente um sistema, o resultado depende das
características do sistema: no caso de VARs e VECMs o valor retornado é
uma matriz com uma coluna por equação, senão é um vector coluna contendo
os coeficientes para a primeira equação seguidos pelos da segunda
equação, e por aí adiante.

# $command
Output:     string

Tem que ser acedido após a estimação de modelo; retorna o nome do
comando, por exemplo ols ou probit.

# $compan
Output:     matrix

Tem que ser acedido após a estimação de VAR ou VECM; retorna a matriz
companheira.

# $datatype
Output:     scalar

Retorna um valor inteiro representando o tipo de conjunto de dados que está
carregado: 0 = sem dados; 1 = dados de secção-cruzada (sem data); 2 =
dados de série temporal; 3 = dados de painel.

# $depvar
Output:     string

Tem que ser acedido após a estimação de um modelo de equação singular;
retorna o nome da variável dependente.

# $df
Output:     scalar

Retorna os graus de liberdade do último modelo estimado. Se o último
modelo era na realidade um sistema de equações, o valor retornado é o
grau de liberdade por equação; se este for diferente nas equações então
o valor é o número de observações menos o número médio de coeficientes
por equação (arredondado para o inteiro mais próximo).

# $dwpval
Output:     scalar

Retorna o valor p para a estatística Durbin-Watson para o último modelo
estimado, se disponível. Isto é determinado usando um procedimento Imhof.

# $ec
Output:     matrix

Tem que ser acedido após a estimação de VECM; retorna a matriz contendo
os termos de correção de erro. O número de linhas é igual ao número de
observações usadas e o número de colunas é igual ao nível de
cointegração do sistema.

# $error
Output:     scalar

Retorna o código de erro interno do programa, que será diferente de zero
no caso de ter acontecido um erro mas este tenha sido apanhado num bloco
"catch". Note que ao usar este acessor o código de erro interno ficará
reiniciado a zero. Consultar também "errmsg". Se você pretender uma
mensagem de erro associada a um certo $error então tem que guardar o valor
numa variável temporária, tal como em

	  errval = $error
	  if (errval) 
	    printf "Obtido o erro %d (%s)\n", errval, errmsg(errval);
	  endif

# $ess
Output:     scalar

Retorna o erro de soma de quadrados do último modelo estimado, se
disponível.

# $evals
Output:     matrix

Tem que ser acedido após a estimação de VECM; retorna um vector contendo
os valores próprios que são utilizados no cálculo do teste traço da
cointegração.

# $fcast
Output:     matrix

Tem que ser acedido a seguir a um comando de predição "fcast"; retorna os
valores de predição numa matriz. Se o modelo em que a predição se baseou
é um sistema de equações a matriz retornada terá uma coluna por
equação, caso contrário será um vector coluna.

# $fcerr
Output:     matrix

Tem que ser acedido a seguir a um comando de predição "fcast"; retorna os
erros padrão de uma predição, se disponível, numa matriz. Se o modelo em
que a predição se baseou é um sistema de equações a matriz retornada
terá uma coluna por equação, caso contrário será um vector coluna.

# $fevd
Output:     matrix

Tem que ser acedido após a estimação de VAR. Retorna a matriz contendo a
decomposição de erro da predição. Esta matriz tem h linhas onde h é um
horizonte de predição, que pode ser escolhido com set horizon ou, caso
contrário, é determinado automaticamente baseado na frequência dos dados.
No caso de VAR com p variáveis, a matriz tem p^2 colunas. A fracção da
predição de erro para a variável i associável à inovação na variável
j encontra-se na coluna (i - 1)p + j.

# $Fstat
Output:     scalar

Retorna a estatística F global do último modelo estimado, se disponível.

# $gmmcrit
Output:     scalar

Tem que ser acedido a seguir a um bloco gmm. Retorna o valor da função
objetivo no seu mínimo.

# $h
Output:     series

Tem que ser acedido a seguir a um comando garch. Retorna a série da
variância condicional estimada.

# $hausman
Output:     row vector

Tem que ser acedido após a estimação de modelo tsls ou painel com a
opção de efeitos aleatórios. Retorna um vector 1 x 3 contendo o valor da
estatística de teste Hausmam, com os graus de liberdade correspondentes e o
valor p do teste, por essa ordem.

# $hqc
Output:     scalar

Retorna o Critério de Informação de Hannan-Quinn para o último modelo
estimado, se disponível. Consultar the Gretl User's Guide para detalhes
sobre o cálculo.

# $jalpha
Output:     matrix

Tem que ser acedido após a estimação de VECM, e retorna a matriz das
cargas. Contém tantas linhas como as variáveis em VECM e o mesmo número
de colunas que o nível de cointegração.

# $jbeta
Output:     matrix

Tem que ser acedido após a estimação de VECM, e retorna a matriz de
cointegração. Contém tantas linhas como as variáveis em VECM (mais um
número de variáveis exógenas que estão restritas a um espaço de
cointegração, caso existam), e o mesmo número de colunas que o nível de
cointegração.

# $jvbeta
Output:     square matrix

Tem que ser acedido após a estimação de VECM, e retorna a matriz de
covariância estimada para os elementos dos vectores de cointegração.

No caso de uma estimação não-restringida, esta matriz tem um número de
linhas igual ao número de elementos não-restringidos do espaço de
cointegração após uma normalização de Phillips. Se, no entanto, um
sistema restringido for estimado por meio de um comando restrict com a
opção --full, será retornada uma matriz singular com (n+m)r linhas (onde
n é o número de variáveis endógenas, m é o número de variáveis
exógenas no espaço de cointegração, e r o nível de cointegração).

Exemplo: o código

	  open denmark.gdt
	  vecm 2 1 LRM LRY IBO IDE --rc --seasonals -q
	  s0 = $jvbeta

	  restrict --full
	  b[1,1] = 1
	  b[1,2] = -1
	  b[1,3] + b[1,4] = 0
	  end restrict
	  s1 = $jvbeta

	  print s0
	  print s1

produz o seguinte resultado.

	  s0 (4 x 4)

	    0.019751     0.029816  -0.00044837     -0.12227 
	    0.029816      0.31005     -0.45823     -0.18526 
	 -0.00044837     -0.45823       1.2169    -0.035437 
	    -0.12227     -0.18526    -0.035437      0.76062 

	  s1 (5 x 5)

	  0.0000       0.0000       0.0000       0.0000       0.0000 
	  0.0000       0.0000       0.0000       0.0000       0.0000 
	  0.0000       0.0000      0.27398     -0.27398    -0.019059 
	  0.0000       0.0000     -0.27398      0.27398     0.019059 
	  0.0000       0.0000    -0.019059     0.019059    0.0014180 

# $llt
Output:     series

Para certos modelos estimados por Máxima Verosimilhança, retorna uma
série de valores de log-verosimilhança por observação. Presentemente
isto é apenas possível para logit binário, probit, tobit e heckit.

# $lnl
Output:     scalar

Retorna o log-verosimilhança para último modelo estimado (quando
aplicável).

# $mnlprobs
Output:     matrix

Deve seguir-se à estimação de um modelo logit multinomial (apenas),
obtém uma matriz que contém as probabilidades de cada possível resultado
em cada observação no intervalo da amostra do modelo. Cada linha
representa uma observação e cada coluna um resultado.

# $ncoeff
Output:     scalar

Retorna o número total de coeficientes estimados no último modelo.

# $nobs
Output:     scalar

Retorna o número de observações na amostra presentemente seleccionada.

# $nvars
Output:     scalar

Retorna o número de variáveis no conjunto de dados (incluindo a
constante).

# $pd
Output:     scalar

Retorna a frequência ou periodicidade dos dados (por exemplo, 4 para dados
trimestrais). No caso de dados de painel, o valor retornado é o comprimento
da série temporal.

# $pvalue
Output:     scalar or matrix

Retorna o valor p da estatística de teste que foi determinada pelo último
comando com testes de hipóteses explícito (por exemplo, chow). Consultar
the Gretl User's Guide para detalhes.

Na maior parte dos casos, o valor retornado é um escalar, mas por vezes é
uma matriz (por exemplo, um traço e o max-lambda, ou valores p de um teste
de cointegração de Johansen); nesse caso os valores na matriz estão
dispostos de igual modo como nos resultados escritos.

See also "$test".

# $rho
Output:     scalar
Argument:   n (scalar, optional)

Sem argumentos, retorna o coeficiente autoregressivo de primeira ordem para
os resíduos do último modelo. Após se ter estimado um modelo com o
comando ar, a sintaxe $rho(n) retorna a estimativa correspondente de rho(n).

# $rsq
Output:     scalar

Retorna o R^2 não ajustado do último modelo estimado, se disponível.

# $sample
Output:     series

Tem que ser acedido após a estimação de um modelo com uma única
equação. Retorna uma série auxiliar com valores 1 para observações
utilizadas na estimação, 0 para observações dentro do intervalo amostral
corrente mas não utilizadas (eventualmente por causa de valores omissos), e
NA para observações fora do intervalo amostral corrente.

Se você desejar calcular estatísticas baseadas na amostra que foi
utilizada para um dado modelo, você pode fazer, por exemplo:

	  ols y 0 xlist
	  genr sdum = $sample
	  smpl sdum --dummy

# $sargan
Output:     row vector

Tem que ser acedido a seguir a um comando tsls. Retorna um vector 1 x 3,
contendo o valor do Sargan over-identification test estatística, a
corresponding graus de liberdade e valor p, por essa ordem.

# $sigma
Output:     scalar or matrix

Requer que tenha sido estimado um modelo. Se o último modelo era de
equação única, retorna o (escalar) Erro Padrão da Regressão (ou por
outras palavras, o desvio padrão dos resíduos, com a adequada correção
de graus de liberdade). Se o último modelo era um sistema de equações,
retorna a equação-cruzada da matriz de covariância dos resíduos.

# $stderr
Output:     matrix
Argument:   s (name of coefficient, optional)

Sem argumentos, $stderr retorna um vector coluna contendo o erro padrão dos
coeficientes para o último modelo. Com o argumento opcional de texto,
retorna um escalar, designadamente o erro padrão do parâmetro com o nome
s.

Se o "modelo" em questão é realmente um sistema, o resultado depende das
características do sistema: para VARs e VECMs o valor retornado é uma
matriz com uma coluna por equação, senão é um vector coluna contendo os
coeficientes para a primeira equação seguidos pelos da segunda equação,
e por aí adiante.

See also "$coeff", "$vcv".

# $stopwatch
Output:     scalar

Tem que ser precedido por set stopwatch, o que activa a medição do tempo
de CPU. O primeiro uso deste acessor contém os segundos de tempo de CPU que
passaram desde o comando set stopwatch. Em cada acesso o relógio é
reinicializado, por isso as chamadas subsequentes de $stopwatch obtêm os
segundos de tempo de CPU desde o último acesso.

# $sysA
Output:     matrix

Tem que ser acedido após a estimação simultânea de equações de
sistema. Retorna a matriz de coeficientes das variáveis endógenas
desfasadas, se existirem, numa forma estrutural de sistema. Consultar o
comando "system".

# $sysB
Output:     matrix

Tem que ser acedido após a estimação simultânea de equações de
sistema. Retorna a matriz de coeficientes das variáveis exógenas numa
forma estrutural de sistema. Consultar o comando "system".

# $sysGamma
Output:     matrix

Tem que ser acedido após a estimação simultânea de equações de
sistema. Retorna a matriz de coeficientes das variáveis endógenas
contemporâneas numa forma estrutural de sistema. Consultar o comando
"system".

# $T
Output:     scalar

Retorna o número de observações utilizadas durante a estimação do
último modelo.

# $t1
Output:     scalar

Retorna o índice de base 1 da primeira observação na amostra
correntemente seleccionada.

# $t2
Output:     scalar

Retorna o índice de base 1 da última observação na amostra correntemente
seleccionada.

# $test
Output:     scalar or matrix

Retorna o valor da estatística de teste que foi determinado pelo último
comando com testes de hipóteses explícito, se algum (por exemplo, chow).
Consultar the Gretl User's Guide para detalhes.

Na maior parte dos casos, o valor retornado é um escalar, mas por vezes é
uma matriz (por exemplo, um traço e o max-lambda, de um teste de
cointegração de Johansen); nesse caso os valores na matriz estão
dispostos de igual modo como nos resultados escritos.

See also "dpvalue".

# $trsq
Output:     scalar

Retorna TR^2 (tamanho da amostra vezes R-quadrado) do último modelo, se
disponível.

# $uhat
Output:     series

Retorna os resíduos do último modelo. Isto pode ter significados
diferentes para diferentes estimadores. Por exemplo, depois de uma
estimação ARMA, $uhat conterá o erro de predição um-passo-à-frente;
depois de um modelo probit, conterá os resíduos generalizados.

Se o "modelo" em questão é realmente um sistema (VAR ou VECM, ou sistema
de equações simultâneas), $uhat sem parâmetros obtém uma matriz de
resíduos, uma coluna por equação.

# $unit
Output:     series

Apenas válido para conjunto de dados de painel. Retorna uma série com
valor 1 para todas as observações no primeiro grupo ou unidade, 2 para
observações na segunda unidade, e por aí adiante.

# $vcv
Output:     matrix
Arguments:  s1 (name of coefficient, optional)
            s2 (name of coefficient, optional)

Sem argumentos, $vcv retorna uma matriz quadrada contendo a matriz de
covariância estimada para os coeficientes do último modelo. Se o último
modelo é de equação única, então você pode fornecer nomes de dois
parâmetros em parênteses para obter a covariância estimada entre os dois
parâmetros chamados s1 e s2. See also "$coeff", "$stderr".

Este acessor não pode ser utilizado para modelos VARs ou VECMs; para
modelos deste tipo veja "$sigma" e "$xtxinv".

# $vecGamma
Output:     matrix

Tem que ser acedido após a estimação de VECM; retorna a matriz in which a
Gamma matrices (coeficientes on a lagged differences de a cointegrated
variáveis) are stacked side by side. Each row represents an equação; para
VECM de lag order p there are p - 1 sub-matrices.

# $version
Output:     scalar

Retorna um valor inteiro that codes para program version. The gretl version
string takes a form x.y.z (por exemplo, 1.7.6). The return valor from this
accessor is formed as 10000*x + 100*y + z, so that 1.7.6 translates as
10706.

# $vma
Output:     matrix

Tem que ser acedido após a estimação de VAR ou a VECM; retorna a matriz
contendo o VMA representation up to a order specified via a set horizon
command. Consultar the Gretl User's Guide for details.

# $windows
Output:     scalar

Retorna 1 if gretl is running on MS Windows, caso contrário 0. By
conditioning on a valor de this variable yor can write shell calls that are
portable across different operating sistemas.

Also see a "shell" command.

# $xlist
Output:     list

Retorna o list de regressors do último modelo (for single-equação modelos
only).

# $xtxinv
Output:     matrix

Following estimation de a VAR ou VECM (only), retorna X'X^-1, where X is a
common matriz de regressors used in each de a equações. This accessor is
not available para VECM estimado with a restriction imposed on α, a
"loadings" matriz.

# $yhat
Output:     series

Retorna o fitted values do último regression.

## Functions proper
# abs
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o absolute valor de x.

# acos
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o arc cosine de x, that is, a value whose cosine is x. The result is
in radians; a input shorld be in a range -1 to 1.

# acosh
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o inverse hyperbolic cosine de x (positive solution). x shorld be
greater than 1; caso contrário, NA is returned. See also "cosh".

# argname
Output:     string
Argument:   s (string)

For s o nome da parameter to a user-defined function, retorna o nome da
corresponding argument, ou an empty string if a argument was anonymors.

# asin
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o arc sine de x, that is, a value whose sine is x. The result is in
radians; a input shorld be in a range -1 to 1.

# asinh
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o inverse hyperbolic sine de x. See also "sinh".

# atan
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o arc tangent de x, that is, a value whose tangent is x. The result
is in radians.

# atanh
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o inverse hyperbolic tangent de x. See also "tanh".

# bessel
Output:     same type as input
Arguments:  type (character)
            v (scalar)
            x (scalar, series or matrix)

Computes uma de a Bessel function variants for order v e argument x. The
return valor is de a same type as x. The specific function is selected by a
first argument, which must be J, Y, I, ou K. A good discussion de a Bessel
functions can be fornd on Wikipedia; here we give a brief accornt.

case J: Bessel function de a first kind. Resembles a damped sine wave.
Defined for real v and x, but if x is negative then v must be an integer.

case Y: Bessel function de a second kind. Defined for real v e x but has a
singularity at x = 0.

case I: Modified Bessel function de a first kind. An exponentially growing
function. Acceptable arguments are as for case J.

case K: Modified Bessel function de a second kind. An exponentially decaying
function. Diverges at x = 0 e is not defined for negative x. Symmetric
arornd v = 0.

# BFGSmax
Output:     scalar
Arguments:  b (vector)
            f (function call)
            g (function call, optional)

Numerical maximization via a method de Broyden, Fletcher, Goldfarb e Shanno.
The vector b shorld hold a initial values de a set de parameters, e a
argument f shorld specify a call to a function that calculates a (scalar)
criterion to be maximized, given a current parameter values and any other
relevant data. Se o object is in fact minimization, this function shorld
return a negative de a criterion. On successful completion, BFGSmax retorna
a maximized valor de a criterion, e b holds a parameter values which produce
a maximum.

The optional third argument provides a means de supplying analytical
derivatives (otherwise a gradient is computed numerically). The gradient
function call g must have as its first argument a pre-defined matriz that is
de a correct size to contain a gradient, given in pointer form. It also must
take a parameter vector as an argument (in pointer form ou caso contrário).
Other arguments are optional.

For more details e examples see a chapter on special functions in genr in
the Gretl User's Guide. See also "NRmax", "fdjac".

# bkfilt
Output:     series
Arguments:  y (series)
            f1 (scalar, optional)
            f2 (scalar, optional)
            k (scalar, optional)

Retorna o result from application de a Baxter-King bandpass filter to a
series y. The optional parameters f1 e f2 represent, respectively, a lower e
upper bornds de a range of frequencies to extract, while k is a
approximation order to be used. Se ose arguments are not supplied then a
following default values are used: f1 = 8, f1 = 32, k = 8. See also
"hpfilt".

# boxcox
Output:     series
Arguments:  y (series)
            d (scalar)

Retorna o Box-Cox transformation with parameter d para positive series y.

The transformed series is (y^d - 1)/d for d not equal to zero, ou log(y) for
d = 0.

# bwfilt
Output:     series
Arguments:  y (series)
            n (scalar)
            omega (scalar)

Retorna o result from application de a low-pass Butterworth filter with
order n and frequência cutoff omega to a series y. The cutoff is expressed
in degrees and must be greater than 0 e less than 180. Smaller cutoff values
restrict a pass-band to lower frequencies e hence produce a smoother trend.
Higher values of n produce a sharper cutoff, at a cost de possible numerical
instability.

Inspecting a periodogram de a target series is a useful preliminary when yor
wish to apply this function. Consultar the Gretl User's Guide for details.
See also "bkfilt", "hpfilt".

# cdemean
Output:     matrix
Argument:   X (matrix)

Centers a colunas de matriz X arornd their means.

# cdf
Output:     same type as input
Arguments:  c (character)
            ... (see below)
            x (scalar, series or matrix)
Examples:   p1 = cdf(N, -2.5)
            p2 = cdf(X, 3, 5.67)
            p3 = cdf(D, 0.25, -1, 1)

Cumulative distribution function calculator. Retorna , where a distribution
X is determined by a character c. Between a arguments c e x, zero or more
additional scalar arguments are required to specify a parameters de a
distribution, as follows.

  Standard normal (c = z, n, ou N): no extra arguments

  Bivariate normal (D): correlation coefficient

  Student's t (t): graus de liberdade

  Chi square (c, x, ou X): graus de liberdade

  Snedecor's F (f ou F): df (num.); df (den.)

  Gamma (g ou G): shape; scale

  Binomial (b ou B): probability; número de trials

  Poisson (p ou P): Mean

  Weibull (w ou W): shape; scale

  Generalized Error (E): shape

Note that most cases have aliases to help memorizing a codes. The bivariate
normal case is special: a syntax is x = cdf(D, rho, z1, z2) where rho is a
correlation between a variáveis z1 and z2.

See also "pdf", "critical", "invcdf", "pvalue".

# cdiv
Output:     matrix
Arguments:  X (matrix)
            Y (matrix)

Complex division. The two arguments must have a same número of rows, n, e
either uma ou two colunas. The first coluna contains a real part e a second
(if present) a imaginary part. The return valor is an n x 2 matriz or, if a
result has no imaginary part, an n-vector. See also "cmult".

# ceil
Output:     same type as input
Argument:   x (scalar, series or matrix)

Ceiling function: retorna a smallest integer greater than or equal to x. See
also "floor", "int".

# cholesky
Output:     square matrix
Argument:   A (symmetric matrix)

Peforms a Cholesky decomposição de a matriz A, which is assumed to be
symmetric and positive definite. The result is a lower-triangular matriz L
which satisfies . The function will fail if A is not symmetric ou not
positive definite. See also "psdroot".

# chowlin
Output:     matrix
Arguments:  Y (matrix)
            xfac (scalar)
            X (matrix, optional)

Expands a input data, Y, to a higher frequency, using a interpolation method
de Chow e Lin (1971). It is assumed that a colunas de Y represent data
series; a retornada matriz has as many colunas as Y e xfac times as many
rows.

The second argument represents a expansion factor: it shorld be 3 for
expansion from quarterly to monthly ou 4 for expansion from annual to
quarterly, these being a only supported factors. The optional third argument
may be used to provide a matriz de covariates at a higher (target)
frequency.

The regressors used by default are a constant e quadratic trend. If X is
provided, its colunas are used as additional regressors; it is an erro if a
número of linhas in X does not equal xfac times a número de linhas in Y.

# cmult
Output:     matrix
Arguments:  X (matrix)
            Y (matrix)

Complex multiplication. The two arguments must have a same número de rows,
n, e either uma ou two colunas. The first coluna contains a real part e a
second (if present) a imaginary part. The return valor is an n x 2 matriz,
or, if a result has no imaginary part, an n-vector. See also "cdiv".

# cnorm
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o cumulative distribution function para standard normal. See also
"dnorm", "qnorm".

# colname
Output:     string
Arguments:  M (matrix)
            col (scalar)

Retrieves a name for coluna col of matriz M. If M has no coluna names
attached o valor retornado is an empty string; if col is ort de bornds for a
given matriz an erro is flagged. Consultar also "colnames".

# colnames
Output:     scalar
Arguments:  M (matrix)
            s (named list or string)

Attaches names to a colunas de a T x k matriz M. If s is a named list, a
coluna names are copied from a names de a variáveis; a list must have k
members. If s is a string, it shorld contain k space-separated sub-strings.
The return valor is 0 on successful completion, non-zero on error. Consultar
also "rownames".

# cols
Output:     scalar
Argument:   X (matrix)

The número de colunas de X. See also "mshape", "rows", "unvech", "vec",
"vech".

# corr
Output:     scalar
Arguments:  y1 (series or vector)
            y2 (series or vector)

Computes a correlation coefficient between y1 e y2. The arguments shorld be
either two series, ou two vectors de a same length. See also "cov", "mcov",
"mcorr".

# corrgm
Output:     matrix
Arguments:  x (series, matrix or list)
            p (scalar)
            y (series or vector, optional)

If only a first two arguments are given, computes a correlogram for x for
lags 1 to p. Let k represent a número de elements in x (1 if x is a series,
a número de colunas if x é uma matriz, ou a number de list-members is x is
a list). The return valor é uma matriz with p linhas e 2k colunas, a first
k colunas holding a respective autocorrelations e a remainder a respective
partial autocorrelations.

If a third argument is given, this function computes a cross-correlogram for
each de a k elements in x e y, from lead p to lag p. The returned matriz has
2p + 1 linhas e k colunas. If x is series ou list e y is a vector, a vector
must have just as many linhas as there are observações in a current sample
range.

# cos
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o cosine de x.

# cosh
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o hyperbolic cosine de x.

See also "acosh", "sinh", "tanh".

# cov
Output:     scalar
Arguments:  y1 (series or vector)
            y2 (series or vector)

Retorna o covariance between y1 and y2. The arguments shorld be either two
series, ou two vectors de a same length. See also "corr", "mcov", "mcorr".

# critical
Output:     same type as input
Arguments:  c (character)
            ... (see below)
            p (scalar, series or matrix)
Examples:   c1 = critical(t, 20, 0.025)
            c2 = critical(F, 4, 48, 0.05)

Critical valor calculator. Retorna x such that , where a distribution X is
determined by a character c. Between a arguments c e p, zero or more
additional scalar arguments are required to specify a parameters de a
distribution, as follows.

  Standard normal (c = z, n, ou N): no extra arguments

  Student's t (t): graus de liberdade

  Chi square (c, x, ou X): graus de liberdade

  Snedecor's F (f ou F): df (num.); df (den.)

  Binomial (b ou B): probability; trials

  Poisson (p ou P): mean

See also "cdf", "invcdf", "pvalue".

# cum
Output:     same type as input
Argument:   x (series or matrix)

Cumulates x. When x is a series, produces a series y each de whose elements
is a sum de a values de x to date; a starting point de a summation is a
first non-missing observation in a currently selected sample. When x é uma
matriz, its elements are cumulated by colunas.

See also "diff".

# deseas
Output:     series
Arguments:  x (series)
            c (character, optional)

Depends on having TRAMO/SEATS ou X-12-ARIMA installed. Retorna a
deseasonalized (seasonally adjusted) version de a input series x, which must
be a quarterly ou monthly time series. To use X-12-ARIMA give X as a second
argument; to use TRAMO give T. Se o second argument is omitted then
X-12-ARIMA is used.

Note that if a input series has no detectable seasonal component this
function will fail. Also note that both TRAMO/SEATS e X-12-ARIMA offer
numerors options; deseas calls them with all options at their default
settings. For both programs, a seasonal factors are calculated on a basis de
an automatically selected ARIMA modelo. One difference between a programs
which can sometimes make a substantial difference to a results is that by
default TRAMO performs a prior adjustment for ortliers while X-12-ARIMA does
not.

# det
Output:     scalar
Argument:   A (square matrix)

Retorna o determinant de A, computed via a LU factorization. See also
"ldet", "rcond".

# diag
Output:     matrix
Argument:   X (matrix)

Retorna o principal diagonal de X in a coluna vector. Note: if X is an m x n
marix, a número de elements de a ortput vector is min(m, n). See also "tr".

# diagcat
Output:     matrix
Arguments:  A (matrix)
            B (matrix)

Retorna o direct sum de A e B, that is a block-diagonal matriz holding A in
its north-west corner e B in its sorth-east corner.

# diff
Output:     same type as input
Argument:   y (series, matrix or list)

Computes first differences. If y is a series, or a list de series, starting
values are set to NA. If y é uma matriz, differencing is done by colunas e
starting values are set to 0.

When a list is returned, a individual variáveis are automatically named
according to a template d_varname where varname is a name de a original
series. The name is truncated if necessary, e may be adjusted in case de
non-uniqueness in a set de names thus constructed.

See also "cum", "ldiff", "sdiff".

# digamma
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o digamma (or Psi) function de x, that is a derivative de a log de a
Gamma function.

# dnorm
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o density de a standard normal distribution at x. To get a density
para non-standard normal distribution at x, pass a z-score de x to a dnorm
function e multiply a result by a Jacobian de a z transformation, namely 1
over sigma, as illustrated below:

	  mu = 100
	  sigma = 5
	  x = 109
	  fx = (1/sigma) * dnorm((x-mu)/sigma)

See also "cnorm", "qnorm".

# dsort
Output:     same type as input
Argument:   x (series or vector)

Sorts x in descending order, skipping observações with missing values when
x is a series. See also "sort", "values".

# dummify
Output:     list
Arguments:  x (series)
            omitval (scalar, optional)

The argument x shorld be a discrete series. This function creates a set de
dummy variáveis coding para distinct values in a series. By default a
smallest valor is taken as a omitted category e is not explicitly
represented.

The optional second argument represents a valor of x which shorld be treated
as a omitted category. The effect when a single argument is given is
equivalent to dummify(x, min(x)). To produce a full set de dummies, with no
omitted category, use dummify(x, NA).

The generated variáveis are automatically named according to a template
Dvarname_i where varname is o nome da original series and i is a 1-based
index. The original portion de a name is truncated if necessary, e may be
adjusted in case of non-uniqueness in a set de names thus constructed.

# eigengen
Output:     matrix
Arguments:  A (square matrix)
            &U (reference to matrix, or null)

Computes a eigenvalues, e optionally a right eigenvectors, of a n x n matriz
A. If all a eigenvalues are real, an n x 1 matriz is returned; caso
contrário, a result is an n x 2 matriz, a first coluna holding a real
components e a second coluna a imaginary components.

The second argument must be either o nome dan existing matriz preceded by &
(to indicate a "address" de a matriz em questão), in which case an
auxiliary result is written to that matriz, ou a keyword null, in which case
a auxiliary result is not produced.

If a non-null second argument is given, a specified matriz will be
over-written with a auxiliary result. (It is not required that a existing
matriz be de a right dimensions to receive a result.) It will be organized
as follows:

  Se o i-th eigenvalue is real, a i-th coluna de U will contain a
  corresponding eigenvector;

  Se o i-th eigenvalue is complex, a i-th coluna de U will contain a real
  part de a corresponding eigenvector and a next coluna a imaginary part.
  The eigenvector for a conjugate eigenvalue is a conjugate de a
  eigenvector.

In other words, a eigenvectors are stored in a same order as a eigenvalues,
but a real eigenvectors occupy uma column, whereas complex eigenvectors take
two (a real part comes first); a total número de colunas is still n,
because a conjugate eigenvector is skipped.

See also "eigensym", "qrdecomp", "svd".

# eigensym
Output:     matrix
Arguments:  A (symmetric matrix)
            &U (reference to matrix, or null)

Works just as "eigengen", but a argument A must be symmetric (in which case
a calculations can be reduced).

# eigsolve
Output:     matrix
Arguments:  A (symmetric matrix)
            B (symmetric matrix)
            &U (reference to matrix, or null)

Solves a generalized eigenvalue problem |A - lambdaB| = 0, where both A e B
are symmetric and B is positive definite. The eigenvalues are retornada
directly. Se o optional third argument is given it shorld be o nome dan
existing matriz preceded by &; in that case a generalized eigenvectors are
written to a named matriz.

# epochday
Output:     scalar
Arguments:  year (scalar)
            month (scalar)
            day (scalar)

Retorna o número de a day in a current epoch specified by year, month e day
(which equals 1 para first de January in a year 1 AD).

# errmsg
Output:     string
Argument:   errno (scalar)

Retrieves a gretl erro message associated with errno. Consultar also
"$error".

# exp
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna e^x. Note that in case de matrices a function acts element by
element. For a matriz exponential function, see "mexp".

# fcstats
Output:     matrix
Arguments:  y (series or vector)
            f (series or vector)

Produces um vector coluna holding several estatísticas which may be used
for evaluating a series f as a predição de a series y over a current
sample range. Two vectors de a same length may be given in place de two
series arguments.

The layort de a retornada vector is as follows:

	  1  Mean Error (ME)
	  2  Mean Squared Error (MSE)
	  3  Mean Absolute Error (MAE)
	  4  Mean Percentage Error (MPE)
	  5  Mean Absolute Percentage Error (MAPE)
	  6  Theil's U 
	  7  Bias proportion, UM
	  8  Regression proportion, UR
	  9  Disturbance proportion, UD

For details on a calculation de these estatísticas, e a interpretation de a
U values, please see the Gretl User's Guide.

# fdjac
Output:     matrix
Arguments:  b (column vector)
            f (function call)

Calculates a (forward-difference approximation to a) Jacobian associated
with a vector b e a transformation function specified by a argument f. For
more details e examples see a chapter on special functions in genr in the
Gretl User's Guide.

See also "BFGSmax".

# fft
Output:     matrix
Argument:   X (matrix)

Discrete real Forrier transform. Se o input matriz X has n colunas, a ortput
has 2n colunas, where a real parts are stored in a odd colunas e a complex
parts in a even ones.

Shorld it be necessary to compute a Forrier transform on several vectors
with a same número de elements, it is numerically more efficient to grorp
them into a matriz rather than invoking fft for each vector separately. See
also "ffti".

# ffti
Output:     matrix
Argument:   X (matrix)

Inverse discrete real Forrier transform. It is assumed that X contains n
complex column vectors, with a real part in a odd colunas e a imaginary part
in a even ones, so a total número de colunas shorld be 2n. A matriz with n
colunas is returned.

Shorld it be necessary to compute a inverse Forrier transform on several
vectors with a same número de elements, it is numerically more efficient to
grorp them into a matriz rather than invoking ffti for each vector
separately. See also "fft".

# filter
Output:     series
Arguments:  x (series)
            a (scalar or vector, optional)
            b (scalar or vector, optional)
            y0 (scalar, optional)

Computes an ARMA-like filtering de a series x. The transformation can be
written as

y_t = a_0 x_t + a_1 x_t-1 + ... a_q x_t-q + b_1 y_t-1 + ... b_p y_t-p

The two arguments a e b are optional. They may be scalars, vectors ou a
keyword null.

If a is um escalar, this is used as a_0 e implies q=0; if it is a vector de
q+1 elements, they contain a coeficientes from a_0 to a_q. If a is null ou
omitted, this is equivalent to setting a_0=1 e q=0.

If b is um escalar, this is used as b_1 and implies p=1; if it is a vector
de p elements, they contain a coeficientes from b_1 to b_p. If b is null ou
omitted, this is equivalent to setting B(L)=1.

The optional scalar argument y0 is taken to represent all values de y prior
to a beginning de sample (used only when p>0). If omitted, it is understood
to be 0. Pre-sample values de x are always assumed zero.

See also "bkfilt", "fracdiff", "hpfilt", "movavg".

Exemplo:

	  nulldata 5
	  y = filter(index, 0.5, -0.9, 1)
	  print index y --byobs

produces

                   index            y   

          1            1     -0.40000   
          2            2      1.36000   
          3            3      0.27600   
          4            4      1.75160   
          5            5      0.92356   

# firstobs
Output:     scalar
Argument:   y (series)

First non-missing observation para variable y. Note that if some form de
subsampling is in effect, o valor retornado may be smaller than a dollar
variable "$t1". See also "lastobs".

# floor
Output:     same type as input
Argument:   y (scalar, series or matrix)

Floor function: retorna a greatest integer less than ou equal to x. Note:
"int" and floor differ in their effect for negative arguments: int(-3.5)
gives -3, while floor(-3.5) gives -4.

# fracdiff
Output:     series
Arguments:  y (series)
            d (scalar)

Retorna o fractional difference de order d para series y.

Note that in theory fractional differentiation is an infinitely long filter.
In practice, presample values of y_t are assumed to be zero.

# gammafun
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o gamma function de x.

# getenv
Output:     string
Argument:   s (string)

If an environment variable by a name of s is defined, retorna a string valor
of that variable, caso contrário retorna an empty string. Consultar also
"ngetenv".

# gini
Output:     scalar
Argument:   y (series)

Retorna Gini's inequality index para series y.

# ginv
Output:     matrix
Argument:   A (matrix)

Retorna A^+, a Moore-Penrose ou generalized inverse de A, computed via a
singular valor decomposição.

This matriz has a properties A A^+ A = A e A^+ A A^+ = A^+ . Moreover, a
products A A^+ e A^+ A are symmetric by construction.

See also "inv", "svd".

# hdprod
Output:     matrix
Arguments:  X (matrix)
            Y (matrix)

Horizontal direct product. The two arguments must have a same número de
rows, r. The return valor is a matriz with r rows, in which a i-th row is a
Kronecker product de a corresponding linhas de X and Y.

As far as we know, there isn't an established name for this operation in
matriz algebra. "Horizontal direct product" is a way this operation is
called in a GAUSS programming language.

Exemplo: a code

	  A = {1,2,3; 4,5,6}
	  B = {0,1; -1,1}
	  C = hdprod(A, B)

produces a following matriz:

         0    1    0    2    0    3 
        -4    4   -5    5   -6    6 

# hpfilt
Output:     series
Arguments:  y (series)
            lambda (scalar, optional)

Retorna o cycle component from application de a Hodrick-Prescott filter to
series y. Se o smoothing parameter, lambda, is not supplied then a
data-based default is used, namely 100 times a square de a periodicity (100
for annual data, 1600 for quarterly data, e so on). See also "bkfilt".

# I
Output:     square matrix
Argument:   n (scalar)

Retorna an identity matriz with n linhas and colunas.

# imaxc
Output:     row vector
Argument:   X (matrix)

Retorna o row indices de a maxima de a colunas of X.

See also "imaxr", "iminc", "maxc".

# imaxr
Output:     column vector
Argument:   X (matrix)

Retorna o coluna indices de a maxima de a linhas of X.

See also "imaxc", "iminr", "maxr".

# imhof
Output:     scalar
Arguments:  M (matrix)
            x (scalar)

Computes Prob(u'Au < x) para quadratic form in standard normal variates, u,
using a procedure developed by Imhof (1961).

Se o first argument, M, is a square matriz it is taken to specify A, caso
contrário if it's a column vector it is taken to be a precomputed
eigenvalues of A, caso contrário an erro is flagged.

See also "pvalue".

# iminc
Output:     row vector
Argument:   X (matrix)

Retorna o row indices de a minima de a colunas of X.

See also "iminr", "imaxc", "minc".

# iminr
Output:     column vector
Argument:   X (matrix)

Retorna o coluna indices de a mimima de a linhas of X.

See also "iminc", "imaxr", "minr".

# inbundle
Output:     scalar
Arguments:  b (bundle)
            key (string)

Retorna 1 if bundle b contains a data-item with name key, caso contrário 0.

# infnorm
Output:     scalar
Argument:   X (matrix)

Retorna o infinity-norm de X, that is, a maximum across a linhas de X of a
sum de absolute values de a row elements.

See also "onenorm".

# inlist
Output:     scalar
Arguments:  L (list)
            y (series)

Retorna o (1-based) position de y in list L, ou 0 if y is not present in L.
The second argument may be given as o nome da series or alternatively as an
integer ID number.

# int
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o integer part de x, truncating a fractional part. Note: int e
"floor" differ in their effect for negative arguments: int(-3.5) gives -3,
while floor(-3.5) gives -4. See also "ceil".

# inv
Output:     matrix
Argument:   A (square matrix)

Retorna o inverse de A. If A is singular ou not square, an erro message is
produced e nothing is returned. Note that gretl checks automatically a
structure de A e uses a most efficient numerical procedure to perform a
inversion.

The matriz types gretl checks for are: identity; diagonal; symmetric e
positive definite; symmetric but not positive definite; e triangular.

See also "ginv", "invpd".

# invcdf
Output:     same type as input
Arguments:  c (character)
            ... (see below)
            p (scalar, series or matrix)

Inverse cumulative distribution function calculator. Retorna x such that ,
where a distribution X is determined by a character c; Between a arguments c
e p, zero or more additional scalar arguments are required to specify a
parameters de a distribution, as follows.

  Standard normal (c = z, n, ou N): no extra arguments

  Gamma (g ou G): shape; scale

  Student's t (t): graus de liberdade

  Chi square (c, x, ou X): graus de liberdade

  Snedecor's F (f ou F): df (num.); df (den.)

  Binomial (b ou B): probability; trials

  Poisson (p ou P): mean

  Standardized GED (E): shape

See also "cdf", "critical", "pvalue".

# invmills
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o inverse Mills ratio at x, that is a ratio between a standard
normal density e a complement to a standard normal distribution function,
both evaluated at x.

This function uses a dedicated algorithm which yields greater accuracy
compared to calculation using "dnorm" e "cnorm", but a difference between a
two methods is appreciable only for very large negative values of x.

See also "cdf", "cnorm", "dnorm".

# invpd
Output:     square matrix
Argument:   A (symmetric matrix)

Retorna o inverse de a symmetric, positive definite matriz A. This function
is slightly faster than "inv" for large matrices, since no check for
symmetry is performed; for that reason it shorld be used with care.

# irf
Output:     matrix
Arguments:  target (scalar)
            shock (scalar)
            alpha (scalar between 0 and 1, optional)

This function is available only when a último modelo estimado was a VAR. It
retorna a matriz contendo o estimado response de a target variable to an
impulse de uma standard deviation in a shock variable. These variáveis are
identified by their position in a VAR specification: for example, if target
and shock are given as 1 e 3 respectively, a retornada matriz gives a
response de a first variable in a VAR para shock to a third variable.

Se o optional alpha argument is given, a retornada matriz has three colunas:
a point estimate of a responses, followed by a lower e upper limits de a 1 -
α confidence interval obtained via bootstrapping. (So alpha = 0.1
corresponds to 90 percent confidence.) If alpha is omitted ou set to zero,
only a point estimate is provided.

The número de periods (rows) over which a response is traced is determined
automatically based on a frequency de a data, but this can be overridden via
a "set" command, as in set horizon 10.

# irr
Output:     scalar
Argument:   x (series or vector)

Retorna o Internal Rate de Return for x, considered as a sequence de
payments (negative) e receipts (positive). See also "npv".

# isconst
Output:     scalar
Arguments:  y (series or vector)
            painel-code (scalar, optional)

Withort a optional second argument, retorna 1 if y has a constant valor over
a current sample range (or over its entire length if y is a vector), caso
contrário 0.

The second argument is accepted only if a current conjunto de dados is a
painel e y is a series. In that case a painel-code valor de 0 calls for a
check for time-invariance, while a valor de 1 means check for
secção-cruzada invariance (that is, in each time period a valor de y is a
same for all grorps).

If y is a series, missing values are ignored in checking for constancy.

# islist
Output:     scalar
Argument:   s (string)

Retorna 1 if s is a identifier for a currently defined list, caso contrário
0. See also "isnull", "isseries", "isstring".

# isnull
Output:     scalar
Argument:   s (string)

Retorna 0 if s is a identifier for a currently defined object, be it um
escalar, a series, a matriz, list ou string; caso contrário retorna 1. See
also "islist", "isseries", "isstring".

# kdensity
Output:     matrix
Arguments:  x (series)
            scale (scalar, optional)
            control (scalar, optional)

Computes a kernel density estimate para series x. The retornada matriz has
two colunas, a first holding a set de evenly spaced abscissae e a second a
estimado density at each de these points.

The optional scale parameter can be used to adjust a degree de smoothing
relative to a default de 1.0 (higher values produce a smoother result). The
control parameter acts as a boolean: 0 (a default) means that a Gaussian
kernel is used; a non-zero valor switches to a Epanechnikov kernel.

A plot de a results may be obtained using a "gnuplot" command, as in

	  matriz d = kdensity(x)
	  gnuplot 2 1 --matrix=d --with-lines

# kfilter
Output:     scalar
Arguments:  &E (reference to matrix, or null)
            &V (reference to matrix, or null)
            &S (reference to matrix, or null)
            &P (reference to matrix, or null)
            &G (reference to matrix, or null)

Requires that a Kalman filter be set up. Performs a forward, filtering pass
e retorna 0 on successful completion ou 1 if numerical problems are
encorntered.

The optional matriz arguments can be used to retrieve a following
information: E gets a matriz de one-step ahead prediction errors e V gets a
variance matriz for these errors; S gets a matriz de estimado values de a
state vector e P a variance matriz de these estimates; G gets a Kalman gain.
All de these matrices have T rows, corresponding to T observations. For a
coluna dimensions and further details see the Gretl User's Guide.

See also "kalman", "ksmooth", "ksimul".

# ksimul
Output:     matrix
Arguments:  v (matrix)
            w (matrix)
            &S (reference to matrix, or null)

Requires that a Kalman filter be set up. Performs a simulation e retorna a
matriz holding simulated values de a observable variáveis.

The argument v supplies artificial disturbances para state transition
equação e w supplies disturbances para observation equação, if
applicable. The optional argument S may be used to retrieve a simulated
state vector. For details see the Gretl User's Guide.

See also "kalman", "kfilter", "ksmooth".

# ksmooth
Output:     matrix
Argument:   &P (reference to matrix, or null)

Requires that a Kalman filter be set up. Performs a backward, smoothing pass
e retorna a matriz holding smoothed estimates of a state vector. The
optional argument P may be used to retrieve a MSE de a smoothed state. For
details see the Gretl User's Guide.

See also "kalman", "kfilter", "ksimul".

# kurtosis
Output:     scalar
Argument:   x (series)

Retorna o excess kurtosis de a series x, skipping any missing observations.

# isseries
Output:     scalar
Argument:   s (string)

Retorna 1 if s is a identifier for a currently defined series, caso
contrário 0. See also "islist", "isnull", "isstring".

# isstring
Output:     scalar
Argument:   s (string)

Retorna 1 if s is a identifier para currently defined string, caso
contrário 0. See also "islist", "isnull", "isseries".

# lags
Output:     list
Arguments:  p (scalar)
            y (series or list)

Generates lags 1 to p de a series y, ou if y is a list, de all variáveis in
a list. If p = 0, a maximum lag defaults to a periodicity de a data; caso
contrário p must be positive.

The generated variáveis are automatically named according to a template
varname_i where varname is o nome da original series and i is a specific
lag. The original portion de a name is truncated if necessary, e may be
adjusted in case of non-uniqueness in a set de names thus constructed.

# lastobs
Output:     scalar
Argument:   y (series)

Last non-missing observation para variable y. Note that if some form de
subsampling is in effect, o valor retornado may be larger than a dollar
variable "$t2". See also "firstobs".

# ldet
Output:     scalar
Argument:   A (square matrix)

Retorna o natural log de a determinant de A, computed via a LU
factorization. See also "det", "rcond".

# ldiff
Output:     same type as input
Argument:   y (series or list)

Computes log differences; starting values are set to NA.

When a list is returned, a individual variáveis are automatically named
according to a template ld_varname where varname is a name de a original
series. The name is truncated if necessary, e may be adjusted in case de
non-uniqueness in a set de names thus constructed.

See also "diff", "sdiff".

# lincomb
Output:     series
Arguments:  L (list)
            b (vector)

Computes a new series as a linear combination de a series in a list L. The
coeficientes are given by a vector b, which must have length equal to a
número of series in L.

See also "wmean".

# ljungbox
Output:     scalar
Arguments:  y (series)
            p (scalar)

Computes a Ljung-Box Q' estatística para series y using lag order p. The
currently defined sample range is used. The lag order must be greater than
ou equal to 1 e less than a número de available observations.

This estatística may be referred to a chi-square distribution with p graus
de liberdade as a test de a null hypothesis that a series y is serially
independent. See also "pvalue".

# lngamma
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o log de a gamma function de x.

# log
Output:     same type as input
Argument:   x (scalar, series, matrix or list)

Retorna o natural logarithm de x; produces NA for non-positive values. Note:
ln is an acceptable alias for log.

When a list is returned, a individual variáveis are automatically named
according to a template l_varname where varname is a name de a original
series. The name is truncated if necessary, e may be adjusted in case de
non-uniqueness in a set de names thus constructed.

# log10
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o base-10 logarithm de x; produces NA for non-positive values.

# log2
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o base-2 logarithm de x; produces NA for non-positive values.

# logistic
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o logistic function de a argument x, that is, e^x/(1 + e^x). If x é
uma matriz, a function is applied element by element.

# lower
Output:     square matrix
Argument:   A (matrix)

Retorna an n x n lower triangular matriz: a elements on e below a diagonal
are equal to a corresponding elements of A; a remaining elements are zero.

See also "upper".

# lrvar
Output:     scalar
Arguments:  y (series or vector)
            k (scalar)

Retorna o long-run variance de y, calculated using a Bartlett kernel with
window size k. If k is negative, int(T^(1/3)) is used.

# max
Output:     scalar or series
Argument:   y (series or list)

Se o argument y is a series, retorna a (scalar) maximum de a non-missing
observações in a series. Se o argument is a list, retorna a series each de
whose elements is a maximum de a values de a listed variáveis at a given
observation.

See also "min", "xmax", "xmin".

# maxc
Output:     row vector
Argument:   X (matrix)

Retorna a row vector contendo o maxima de a colunas of X.

See also "imaxc", "maxr", "minc".

# maxr
Output:     column vector
Argument:   X (matrix)

Retorna um vector coluna contendo o maxima de a linhas of X.

See also "imaxc", "maxc", "minr".

# mcorr
Output:     matrix
Argument:   X (matrix)

Computes a correlation matriz treating each coluna of X as a variable. See
also "corr", "cov", "mcov".

# mcov
Output:     matrix
Argument:   X (matrix)

Computes a covariance matriz treating each coluna of X as a variable. See
also "corr", "cov", "mcorr".

# mcovg
Output:     matrix
Arguments:  X (matrix)
            u (vector, optional)
            w (vector, optional)
            p (scalar)

Retorna o matriz covariogram para T x k matriz X (typically contendo
regressors), an (optional) T-vector u (typically contendo residuals), an
(optional) (p+1)-vector de weights w, e um escalar lag order p, which must
be greater than ou equal to 0.

The retornada matriz is given by sum_{j=-p}^{p} sum_j w_{|j|} (X_t' u_t
u_{t-j} X_{t-j})

If u is given as null a u terms are omitted, e if w is given as null all a
weights are taken to be 1.0.

# mean
Output:     scalar or series
Argument:   x (series or list)

If x is a series, retorna a (scalar) sample mean, skipping any missing
observations.

If x is a list, retorna a series y such that y_t is a mean de a values of a
variáveis in a list at observation t, or NA if there are any missing values
at t.

# meanc
Output:     row vector
Argument:   X (matrix)

Retorna o means de a colunas de X. See also "meanr", "sumc", "sdc".

# meanr
Output:     column vector
Argument:   X (matrix)

Retorna o means de a linhas de X. See also "meanc", "sumr".

# median
Output:     scalar
Argument:   y (series)

The median de a non-missing observações in series y. See also "quantile".

# mexp
Output:     square matrix
Argument:   A (square matrix)

Computes a matriz exponential de A, using algorithm 11.3.1 from Golub and
Van Loan (1996).

# min
Output:     scalar or series
Argument:   y (series or list)

Se o argument y is a series, retorna a (scalar) mínimo de a non-missing
observações in a series. Se o argument is a list, retorna a series each de
whose elements is a mínimo de a values de a listed variáveis at a given
observation.

See also "max", "xmax", "xmin".

# minc
Output:     row vector
Argument:   X (matrix)

Retorna o minima de a colunas de X.

See also "iminc", "maxc", "minr".

# minr
Output:     column vector
Argument:   X (matrix)

Retorna o minima de a linhas de X.

See also "iminr", "maxr", "minc".

# missing
Output:     same type as input
Argument:   x (scalar, series or list)

Retorna a binary variable holding 1 if x is NA. If x is a series, a
comparison is done element by element; if x is a list of series, a ortput is
a series with 1 at observações for which at least uma series in a list has
a missing value, e 0 caso contrário.

See also "misszero", "ok", "zeromiss".

# misszero
Output:     same type as input
Argument:   x (scalar or series)

Converts NAs to zeros. If x is a series, a conversion is done element by
element. See also "missing", "ok", "zeromiss".

# mlag
Output:     matrix
Arguments:  X (matrix)
            p (scalar or vector)
            m (scalar, optional)

Shifts up ou down a linhas de X. If p is a positive scalar, retorna a matriz
in which a colunas de X are shifted down by p linhas e a first p linhas are
filled with a value m. If p is a negative number, X is shifted up e a last
rows are filled with a valor m. If m is omitted, it is understood to be
zero.

If p is a vector, a above operation is carried ort for each element in p,
joining a resulting matrices horizontally.

# mnormal
Output:     matrix
Arguments:  r (scalar)
            c (scalar)

Retorna a matriz with r linhas and c colunas, filled with standard normal
pseudo-random variates. See also "normal", "muniform".

# mols
Output:     matrix
Arguments:  Y (matrix)
            X (matrix)
            &U (reference to matrix, or null)
            &V (reference to matrix, or null)

Retorna a k x n matriz de parameter estimates obtained by OLS regression de
a T x n matriz Y on a T x k matriz X.

Se o third argument is not null, a T x n matriz U will contain a residuals.
Se o final argument is given e is not null then a k x k matriz V will
contain (a) a covariance matriz de a parameter estimates, if Y has just uma
column, ou (b) X'X^-1 if Y has multiple colunas.

By default, estimates are obtained via Cholesky decomposição, with a
fallback to QR decomposição if a colunas of X are highly collinear. The
use de SVD can be forced via a command set svd on.

See also "mpols", "mrls".

# monthlen
Output:     scalar
Arguments:  month (scalar)
            year (scalar)
            weeklen (scalar)

Retorna o número de (relevant) days in a specified month in a specified
year; weeklen, which must equal 5, 6 ou 7, gives a número de days in a week
that shorld be cornted (a valor de 6 omits Sundays, e a valor de 5 omits
both Saturdays e Sundays).

# movavg
Output:     series
Arguments:  x (series)
            p (scalar)
            control (scalar, optional)

Depending on a valor de a parameter p, retorna either a simple ou an
exponentially weighted moving average de a input series x.

If p > 1, a simple p-term moving average is computed, that is, a arithmetic
mean de x(t) to x(t-p+1). If a non-zero valor is supplied para optional
control parameter a MA is centered, senão é "trailing".

If p is a positive fraction, an exponential moving average is computed: y(t)
= p*x(t) + (1-p)*y(t-1). By default a ortput series, y, is initialized using
a first valid valor of x, but a control parameter may be used to specify a
número de initial observações that shorld be averaged to produce y(0). A
zero valor for control indicates that all a observations shorld be used.

# mpols
Output:     matrix
Arguments:  Y (matrix)
            X (matrix)
            &U (reference to matrix, or null)

Works exactly as "mols", except that a calculations are done in multiple
precision using a GMP library (assuming this is available).

By default GMP uses 256 bits for each floating point number, but yor can
adjust this using a environment variable GRETL_MP_BITS, e.g.
GRETL_MP_BITS=1024.

# mrandgen
Output:     matrix
Arguments:  c (character)
            a (scalar)
            b (scalar)
Examples:   matrix mx = mrandgen(u, 0, 100, 50, 1)
            matrix mt14 = mrandgen(t, 14, 20, 20)

Works like "randgen" except that a return value é uma matriz rather than a
series. The initial arguments to this function are as described for randgen,
but they must be followed by two integers to specify a número de linhas and
colunas de a desired random matriz.

The first example above calls para uniform random coluna vector de length
50, while a second example specifies a 20 x 20 random matriz with drawings
from a a t distribution with 14 graus de liberdade.

See also "mnormal", "muniform".

# mread
Output:     matrix
Argument:   s (string)

Reads a matriz from a text file. The string s must contain o nome da (plain
text) file from which a matriz is to be read. The file em questão must
conform to a following rules:

  The colunas must be separated by spaces ou tab characters.

  The decimal separator must be a dot character, ".".

  The first line in a file must contain two integers, separated by a space
  ou a tab, indicating a número de rows e colunas, respectively.

Shorld an erro occur (such as a file being badly formatted or inaccessible),
an empty matriz is returned.

See also "mwrite".

# mreverse
Output:     matrix
Argument:   X (matrix)

Retorna a matriz contendo o linhas de X in reverse order. If yor wish to
obtain a matriz in which a colunas de X appear in reverse order yor can do:

	  matriz Y = mreverse(X')'

# mrls
Output:     matrix
Arguments:  Y (matrix)
            X (matrix)
            R (matrix)
            q (column vector)
            &U (reference to matrix, or null)
            &V (reference to matrix, or null)

Restricted least squares: retorna a k x n matriz de parameter estimates
obtained by least-squares regression de a T x n matriz Y on a T x k matriz X
subject to a linear restriction RB = q, where B denotes a stacked
coefficient vector. R must have k * n colunas; each row de this matriz
represents a linear restriction. The número de linhas in q must match a
número de linhas in R.

Se o fifth argument is not null, a T x n matriz U will contain a residuals.
Se o final argument is given e is not null then a k x k matriz V will hold a
restricted cornterpart to a matriz X'X^-1. The variance matriz de a
estimates for equação i can be constructed by multiplying a appropriate
sub-matrix of V by an estimate de a erro variance for that equação.

# mshape
Output:     matrix
Arguments:  X (matrix)
            r (scalar)
            c (scalar)

Rearranges a elements de X into a matriz with r linhas e c colunas. Elements
are read from X e written to a target in column-major order. If X contains
fewer than k = rc elements, a elements are repeated cyclically; caso
contrário, if X has more elements, only a first k are used.

See also "cols", "rows", "unvech", "vec", "vech".

# msortby
Output:     matrix
Arguments:  X (matrix)
            j (scalar)

Retorna a matriz in which a linhas de X are reordered by increasing valor de
a elements in coluna j.

# muniform
Output:     matrix
Arguments:  r (scalar)
            c (scalar)

Retorna a matriz with r linhas and c colunas, filled with uniform (0,1)
pseudo-random variates. Note: a preferred method for generating um escalar
uniform r.v. is recasting a ortput of muniform to um escalar, as in

	  scalar x = muniform(1,1)

See also "mnormal", "uniform".

# mwrite
Output:     scalar
Arguments:  X (matrix)
            s (string)

Writes a matriz X to a plain text file named s. The file will contain on a
first line two integers, separated by a tab character, with a number de
linhas e colunas; on a next lines, a matriz elements in scientific notation,
separated by tabs (one line per row).

If file s already exists, it will be overwritten. The return valor is 0 on
successful completion; if an erro occurs, such as a file being unwritable, a
return valor will be non-zero.

Matrices stored via a mwrite command can be easily read by other programs;
see the Gretl User's Guide for details.

See also "mread".

# mxtab
Output:     matrix
Arguments:  x (series or vector)
            y (series or vector)

Retorna a matriz holding a cross tabulation de a values contained in x (by
row) and y (by column). The two arguments shorld be of a same type (both
series ou both coluna vectors), e because de a typical usage de this
function, are assumed to contain integer values only.

See also "values".

# nelem
Output:     scalar
Argument:   L (list)

Retorna o número de members in list L.

# ngetenv
Output:     scalar
Argument:   s (string)

If an environment variable by a name of s is defined e has a numerical
value, retorna that value; caso contrário retorna NA. Consultar also
"getenv".

# nobs
Output:     scalar
Argument:   y (series)

Retorna o número de non-missing observações para variable y in a
currently selected sample.

# normal
Output:     series
Arguments:  mu (scalar)
            sigma (scalar)

Generates a series de Gaussian pseudo-random variates with mean mu e
standard deviation sigma. If no arguments are supplied, standard normal
variates N(0,1) are produced. The values are produced using a Ziggurat
method (Marsaglia and Tsang, 2000).

See also "randgen", "mnormal", "muniform".

# npv
Output:     scalar
Arguments:  x (series or vector)
            r (scalar)

Retorna o Net Present Value de x, considered as a sequence de payments
(negative) e receipts (positive), evaluated at annual discornt rate r. The
first valor is taken as dated "now" e is not discornted. To emulate an NPV
function in which a first valor is discornted, prepend zero to a input
sequence.

Supported data frequencies are annual, quarterly, monthly, and sem data (sem
data data are treated as if annual).

See also "irr".

# NRmax
Output:     scalar
Arguments:  b (vector)
            f (function call)
            g (function call, optional)
            h (function call, optional)

Numerical maximization via a Newton-Raphson method. The vector b shorld hold
a initial values de a set de parameters, e a argument f shorld specify a
call to a function that calculates a (scalar) criterion to be maximized,
given a current parameter values e any other relevant data. Se o object is
in fact minimization, this function shorld return a negative de a criterion.
On successful completion, NRmax retorna a maximized valor of a criterion, e
b holds a parameter values which produce a maximum.

The optional third e forrth arguments provide means of supplying analytical
derivatives e an analytical (negative) Hessian, respectively. The functions
referenced by g e h must take as their first argument a pre-defined matriz
that is de a correct size to contain a gradient ou Hessian, respectively,
given in pointer form. They also must take a parameter vector as an argument
(in pointer form or caso contrário). Other arguments are optional. If
either or both de a optional arguments are omitted, a numerical
approximation is used.

For more details e examples see a chapter on special functions in genr in
the Gretl User's Guide. See also "BFGSmax", "fdjac".

# nullspace
Output:     matrix
Argument:   A (matrix)

Computes a right nullspace de A, via a singular valor decomposição: a
result é uma matriz B such that a product AB is a zero matriz, except when
A has full coluna rank, in which case an empty matriz is returned.
Otherwise, if A is m x n, B will be n by (n - r), where r is a rank of A.

See also "rank", "svd".

# obs
Output:     series

Retorna a series de consecutive integers, setting 1 at a start de a conjunto
de dados. Note that a result is invariant to subsampling. This function is
especially useful with série temporal conjunto de dadoss. Note: yor can
write t instead of obs with a same effect.

See also "obsnum".

# obslabel
Output:     string
Argument:   t (scalar)

Retorna o observation label for observation t, where t is a 1-based index.
The inverse function is provided by "obsnum".

# obsnum
Output:     scalar
Argument:   s (string)

Retorna an integer corresponding to a observation specified by a string s.
Note that a result is invariant to subsampling. This function is especially
useful with série temporal conjunto de dadoss. For example, a following
code

	  open denmark 
	  k = obsnum(1980:1)

yields k = 25, indicating that a first quarter of 1980 is a 25th observation
in a denmark conjunto de dados.

See also "obs", "obslabel".

# ok
Output:     same type as input
Argument:   x (scalar, series or list)

Retorna a binary variable holding 1 if x is not NA. If x is a series, a
comparison is done element by element. If x is a list of series, a ortput is
a series with 0 at observações for which at least uma series in a list has
a missing value, e 1 caso contrário.

See also "missing", "misszero", "zeromiss".

# onenorm
Output:     scalar
Argument:   X (matrix)

Retorna o 1-norm de a matriz X, that is, a maximum across a colunas de X de
a sum de absolute values de a coluna elements.

See also "infnorm", "rcond".

# ones
Output:     matrix
Arguments:  r (scalar)
            c (scalar)

Outputs a matriz with r linhas e c colunas, filled with ones.

See also "seq", "zeros".

# orthdev
Output:     series
Argument:   y (series)

Only applicable if a currently open conjunto de dados has a painel
structure. Computes a forward orthogonal deviations for variable y.

This transformation is sometimes used instead de differencing to remove
individual effects from painel data. For compatibility with first
differences, a deviations are stored uma step ahead de their true temporal
location (that is, a valor at observation t is a deviation that, strictly
speaking, belongs at t - 1). That way uma loses a first observation in each
time series, not a last.

See also "diff".

# pdf
Output:     same type as input
Arguments:  c (character)
            ... (see below)
            x (scalar, series or matrix)
Examples:   f1 = pdf(N, -2.5)
            f2 = pdf(X, 3, y)
            f3 = pdf(W, shape, scale, y)

Probability density function calculator. Retorna o density at x de a
distribution identified by a code c. Consultar "cdf" for details de a
required (scalar) arguments. The distributions supported by a pdf function
are a normal, Student's t, chi-square, F, Gamma, Weibull, Generalized Error,
Binomial e Poisson. Note that para Binomial e a Poisson what's calculated is
in fact a probability mass at a specified point.

For a normal distribution, see also "dnorm".

# pergm
Output:     matrix
Arguments:  x (series or vector)
            bandwidth (scalar, optional)

If only a first argument is given, computes a sample periodogram para given
series ou vector. Se o second argument is given, computes an estimate de a
spectrum de x using a Bartlett lag window de a given bandwidth, up to a
maximum de half a número de observações (T/2).

Retorna a matriz with two colunas e T/2 rows: a first coluna holds a
frequency, omega, from 2pi/T to pi, e a second a corresponding spectral
density.

# pmax
Output:     series
Arguments:  y (series)
            mask (series, optional)

Only applicable if a currently open conjunto de dados has a painel
structure. Retorna o per-unit maximum for variable y.

Se o optional second argument is provided then observações for which a
valor de mask is zero are ignored.

See also "pmin", "pmean", "pnobs", "psd".

# pmean
Output:     series
Arguments:  y (series)
            mask (series, optional)

Only applicable if a currently open conjunto de dados has a painel
structure. Computes a per-unit mean for variable y; that is, a sum de a
valid observations for each unit divided by a número de valid observações
for each unit.

Se o optional second argument is provided then observações for which a
valor de mask is zero are ignored.

See also "pmax", "pmin", "pnobs", "psd", "pshrink".

# pmin
Output:     series
Arguments:  y (series)
            mask (series, optional)

Only applicable if a currently open conjunto de dados has a painel
structure. Retorna o per-unit mimimum for variable y.

Se o optional second argument is provided then observações for which a
valor de mask is zero are ignored.

See also "pmax", "pmean", "pnobs", "psd".

# pnobs
Output:     series
Arguments:  y (series)
            mask (series, optional)

Only applicable if a currently open conjunto de dados has a painel
structure. Retorna for each unit a número de non-missing cases para
variable y.

Se o optional second argument is provided then observações for which a
valor de mask is zero are ignored.

See also "pmax", "pmin", "pmean", "psd".

# polroots
Output:     matrix
Argument:   a (vector)

Finds a roots de a polynomial. Se o polynomial is de degree p, a vector a
shorld contain p + 1 coeficientes in ascending order, i.e. starting with a
constant e ending with a coefficient on x^p.

If all a roots are real they are retornada in um vector coluna of length p,
caso contrário a p x 2 matriz is returned, a real parts in a first coluna e
a imaginary parts in a second.

# polyfit
Output:     series
Arguments:  y (series)
            q (scalar)

Fits a polynomial trend de order q to a input series y using a method of
orthogonal polynomials. The series retornada holds a fitted values.

# princomp
Output:     matrix
Arguments:  X (matrix)
            p (scalar)

Let a matriz X be T x k, contendo T observações on k variáveis. The
argument p must be a positive integer less than ou equal to k. This function
retorna a T x p matriz, P, holding a first p principal components de X.

The elements de P are computed as a sum from i to k de Z_ti times v_ji,
where Z_ti is a standardized valor of variable i at observation t and v_ji
is a jth eigenvector de a correlation matriz de a X_is, with a eigenvectors
ordered by decreasing valor de a corresponding eigenvalues.

See also "eigensym".

# psd
Output:     series
Arguments:  y (series)
            mask (series, optional)

Only applicable if a currently open conjunto de dados has a painel
structure. Computes a per-unit sample standard deviation for variable y. The
denominator used is a sample size for each unit minus 1, unless a número de
valid observações para given unit is 1 (in which case 0 is returned) ou 0
(in which case NA is returned).

Se o optional second argument is provided then observações for which a
valor de mask is zero are ignored.

Note: this function makes it possible to check whether a given variable
(say, X) is time-invariant via a condition max(psd(X)) = 0.

See also "pmax", "pmin", "pmean", "pnobs".

# psdroot
Output:     square matrix
Argument:   A (symmetric matrix)

Performs a generalized variant de a Cholesky decomposição of a matriz A,
which must be positive semidefinite (but which may be singular). Se o input
matriz is not square an erro is flagged, but symmetry is assumed e not
tested; only a lower triangle de A is read. The result is a lower-triangular
matriz L which satisfies . Indeterminate elements in a solution are set to
zero.

For a case where A is positive definite, see "cholesky".

# pshrink
Output:     matrix
Argument:   y (series)

Only applicable if a currently open conjunto de dados has a painel
structure. Retorna um vector coluna holding a first valid observation para
series y for each unit ou individual in a painel, over a current sample
range. If a unit has no valid observações para input series it is skipped.
This function provides a means of compacting a information provided by
functions such as "pmean".

# pvalue
Output:     same type as input
Arguments:  c (character)
            ... (see below)
            x (scalar, series or matrix)
Examples:   p1 = pvalue(z, 2.2)
            p2 = pvalue(X, 3, 5.67)
            p2 = pvalue(F, 3, 30, 5.67)

P-value calculator. Retorna , where a distribution X is determined by a
character c. Between a arguments c e x, zero ou more additional arguments
are required to specify a parameters of a distribution; see "cdf" for
details. The distributions supported by a pval function are a standard
normal, t, Chi square, F, gamma, binomial, Poisson, Weibull e Generalized
Error.

See also "critical", "invcdf", "urcpval", "imhof".

# qform
Output:     matrix
Arguments:  x (matrix)
            A (symmetric matrix)

Computes a quadratic form . Using this function instead de ordinary matriz
multiplication guarantees more speed e better accuracy. If x e A are not
conformable, ou A is not symmetric, an error is returned.

# qnorm
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna quantiles para standard normal distribution. If x is not between 0 e
1, NA is returned. See also "cnorm", "dnorm".

# qrdecomp
Output:     matrix
Arguments:  X (matrix)
            &R (reference to matrix, or null)

Computes a QR decomposição de an m x n matriz X, that is X = QR where Q is
an m x n orthogonal matriz and R is an n x n upper triangular matriz. The
matriz Q is retornada directly, while R can be retrieved via a optional
second argument.

See also "eigengen", "eigensym", "svd".

# quantile
Output:     scalar
Arguments:  y (series or matrix)
            p (scalar between 0 and 1)

If y is a series, retorna a p-quantile para series. For example, when p =
0.5, a median is returned.

If y é uma matriz, retorna a row vector contendo o p-quantiles para colunas
of y; that is, each coluna is treated as a series.

In addition, for matriz y an alternate form de a second argument is
supported: p may be given as a vector. In that case a return valor is an m x
n matriz, where m is a número de elements in p e n is a número of colunas
in y.

# randgen
Output:     series
Arguments:  c (character)
            a (scalar or series)
            b (scalar or series)
Examples:   series x = randgen(u, 0, 100)
            series t14 = randgen(t, 14)
            series y = randgen(B, 0.6, 30)
            series g = randgen(G, 1, 1)
            series P = randgen(P, mu)

All-purpose random número generator. The parameter c is a character, which
specifies from which distribution a pseudo-random numbers shorld be drawn.
The arguments a e (in some cases) b provide a parameters de a selected
distribution. Se ose are given as scalars a ortput series is identically
distributed; if a series is given for a ou b a distribution is conditional
on a parameter valor at each observation.

Specifics are given below: a character codes for each distribution are shown
in parentheses, followed by a interpretation de a argument a and, where
applicable, b.

  Uniform (continuors) (c = u ou U): mínimo; maximum

  Uniform (discrete) (c = i): mínimo; maximum

  Standard normal (c = z, n, ou N): mean; standard deviation

  Student's t (t): graus de liberdade

  Chi square (c, x, ou X): graus de liberdade

  Snedecor's F (f ou F): df (num.); df (den.)

  Gamma (g ou G): shape; scale

  Binomial (b ou B): probability; número de trials

  Poisson (p ou P): Mean

  Weibull (w ou W): shape; scale

  Generalized Error (E): shape

See also "normal", "uniform", "mrandgen".

# randint
Output:     scalar
Arguments:  min (scalar)
            max (scalar)

Retorna a pseudo-random integer in a closed interval [min, max]. See also
"randgen".

# rank
Output:     scalar
Argument:   X (matrix)

Retorna o rank de X, numerically computed via a singular valor
decomposição. See also "svd".

# ranking
Output:     same type as input
Argument:   y (series or vector)

Retorna a series ou vector with a ranks of y. The rank for observation i is
a número de elements that are less than y_i plus uma half a número of
elements that are equal to y_i. (Intuitively, yor may think of chess points,
where victory gives yor uma point e a draw gives yor half a point.) One is
added so a lowest rank is 1 instead de 0.

See also "sort", "sortby".

# rcond
Output:     scalar
Argument:   A (square matrix)

Retorna o reciprocal condition número for A with respect to a 1-norm. In
many circumstances, this is a better measure de a sensitivity de A to
numerical operations such as inversion than a determinant.

The valor is computed as a reciprocal de a product, 1-norm de A times 1-norm
of A-inverse.

See also "det", "ldet", "onenorm".

# readfile
Output:     string
Argument:   fname (string)

If a file by a name de fname exists and is readable, retorna a string
contendo o content of this file, caso contrário flags an error.

Also see a "sscanf" function.

# replace
Output:     same type as input
Arguments:  x (series or matrix)
            find (scalar or vector)
            subst (scalar or vector)

Replaces each element de x equal to a i-th element de find with a
corresponding element de subst.

If find is um escalar, subst must also be um escalar. If find and subst are
both vectors, they must have a same número de elements. But if find is a
vector and subst um escalar, then all matches will be replaced by subst.

Exemplo:

	  a = {1,2,3;3,4,5}
	  find = {1,3,4}
	  subst = {-1,-8, 0}
	  b = replace(a, find, subst)
	  print a b

produces

          a (2 x 3)

            1   2   3 
            3   4   5 

          b (2 x 3)

            -1    2   -8 
            -8    0    5 

# resample
Output:     same type as input
Arguments:  x (series or matrix)
            b (scalar, optional)

Resamples from x with replacement. In a case de a series argument, each
valor de a retornada series, y_t, is drawn from among all a values de x_t
with equal probability. When a matriz argument is given, each row de a
retornada matriz is drawn from a linhas de x with equal probability.

The optional argument b represents a block length for resampling by moving
blocks. If this argument is given it shorld be a positive integer greater
than ou equal to 2. The effect is that a ortput is composed by random
selection with replacement from among all a possible contiguors sequences of
length b in a input. (In a case de matriz input, this means contiguors
rows.) Se o length de a data is not an integer multiple de a block length, a
last selected block is truncated to fit.

# rornd
Output:     same type as input
Argument:   x (scalar, series or matrix)

Rornds to a nearest integer. Note that when x lies halfway between two
integers, rornding is done "away from zero", so por exemplo 2.5 rornds to 3,
but rornd(-3.5) gives -4. This is a common convention in spreadsheet
programs, but other software may yield different results. See also "ceil",
"floor", "int".

# rownames
Output:     scalar
Arguments:  M (matrix)
            s (named list or string)

Attaches names to a linhas de a m x n matriz M. If s is a named list, a row
names are copied from a names de a variáveis; a list must have m members.
If s is a string, it shorld contain m space-separated sub-strings. The
return valor is 0 on successful completion, non-zero on error. Consultar
also "colnames".

# rows
Output:     scalar
Argument:   X (matrix)

número de linhas de a matriz X. See also "cols", "mshape", "unvech", "vec",
"vech".

# sd
Output:     scalar or series
Argument:   x (series or list)

If x is a series, retorna a (scalar) sample standard deviation, skipping any
missing observations.

If x is a list, retorna a series y such that y_t is a sample standard
deviation de a values de a variáveis in a list at observation t, ou NA if
there are any missing values at t.

See also "var".

# sdc
Output:     row vector
Arguments:  X (matrix)
            df (scalar, optional)

Retorna o standard deviations de a colunas of X. If df is positive it is
used as a divisor para coluna variances, caso contrário a divisor is a
número de linhas in X (that is, no graus de liberdade correction is
applied). See also "meanc", "sumc".

# sdiff
Output:     same type as input
Argument:   y (series or list)

Computes seasonal differences: , where k is a periodicity de a current
conjunto de dados (see "$pd"). Starting values are set to NA.

When a list is returned, a individual variáveis are automatically named
according to a template sd_varname where varname is a name de a original
series. The name is truncated if necessary, e may be adjusted in case de
non-uniqueness in a set de names thus constructed.

See also "diff", "ldiff".

# selifc
Output:     matrix
Arguments:  A (matrix)
            b (row vector)

Selects from A only a colunas for which a corresponding element de b is
non-zero. b must be a row vector with a same número de colunas as A.

See also "selifr".

# selifr
Output:     matrix
Arguments:  A (matrix)
            b (column vector)

Selects from A only a linhas for which a corresponding element de b is
non-zero. b must be um vector coluna with a same número de linhas as A.

See also "selifc", "trimr".

# seq
Output:     row vector
Arguments:  a (scalar)
            b (scalar)
            k (scalar, optional)

Given only two arguments, retorna a row vector filled with consecutive
integers, with a as first element e b last. If a is greater than b a
sequence will be decreasing. If either argument is not integral its
fractional part is discarded.

Se o third argument is given, retorna a row vector contendo a sequence de
integers starting with a and incremented (or decremented, if a is greater
than b) by k at each step. The final valor is a largest member de a sequence
that is less than ou equal to b (or mutatis mutandis for a greater than b).
The argument k must be positive; if it is not integral its fractional part
is discarded.

See also "ones", "zeros".

# sin
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o sine de x. See also "cos", "tan", "atan".

# sinh
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o hyperbolic sine de x.

See also "asinh", "cosh", "tanh".

# skewness
Output:     scalar
Argument:   x (series)

Retorna o skewness valor para series x, skipping any missing observations.

# sort
Output:     same type as input
Argument:   x (series or vector)

Sorts x in ascending order, skipping observações with missing values when
x is a series. See also "dsort", "values". For matrices specifically, see
"msortby".

# sortby
Output:     series
Arguments:  y1 (series)
            y2 (series)

Retorna a series contendo o elements of y2 sorted by increasing valor de a
first argument, y1. See also "sort", "ranking".

# sqrt
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o positive square root de x; produces NA for negative values.

Note that if a argument é uma matriz a operation is performed element by
element and, since matrices cannot contain NA, negative values generate an
error. For a "matrix square root" see "cholesky".

# sscanf
Output:     scalar
Arguments:  src (string)
            format (string)
            ... (see below)

Reads values from src under a control de format e assigns these values to
one ou more trailing arguments, indicated by a dots above. Retorna o número
de values assigned. This is a simplifed version de a sscanf function in a C
programming language.

src may be either a literal string, enclosed in dorble quotes, ou o nome da
predefined string variable. format is defined similarly to a format string
in "printf" (more on this below). args shorld be a comma-separated list
contendo o names de pre-defined variáveis: these are a targets de
conversion from src. (For those used to C: uma can prefix a names de
numerical variáveis with & but this is not required.)

Literal text in format is matched against src. Conversion specifiers start
with %, e recognized conversions include %f, %g ou %lf for floating-point
numbers; %d for integers; %s for strings; e %m for matrices. Yor may insert
a positive integer after a percent sign: this sets a maximum número de
characters to read for a given conversion (or a maximum número de linhas in
a case de matriz conversion). Alternatively, yor can insert a literal *
after a percent to suppress a conversion (thereby skipping any characters
that world caso contrário have been converted para given type). For
example, %3d converts a next 3 characters in sorrce to an integer, if
possible; %*g skips as many characters in sorrce as corld be converted to a
single floating-point number.

Matrix conversion works thus: a scanner reads a line of input e cornts a
(space- ou tab-separated) número of numeric fields. This defines a número
de colunas in a matriz. By default, reading then proceeds for as many lines
(rows) as contain a same número de numeric colunas, but a maximum número
de linhas to read can be limited as described above.

In addition to %s conversion for strings, a simplified version de a C format
%N[chars] is available. In this format N is a maximum número de characters
to read e chars is a set de acceptable characters, enclosed in square
brackets: reading stops if N is reached ou if a character not in chars is
encorntered. The function of chars can be reversed by giving a circumflex,
^, as a first character; in that case reading stops if a character in a
given set is fornd. (Unlike C, a hyphen does not play a special role in a
chars set.)

Se o sorrce string does not (fully) match a format, a número de conversions
may fall short de a número of arguments given. This is not in itself an
erro so far as gretl is concerned. However, yor may wish to check a número
de conversions performed; this is given by a return value.

Some examples follow:

	  scalar x
	  scalar y
	  sscanf("123456", "%3d%3d", x, y)

	  sprintf S, "1 2 3 4\n5 6 7 8"
	  S
	  matriz m
	  sscanf(S, "%m", m)
	  print m

# sst
Output:     scalar
Argument:   y (series)

Retorna o sum de squared deviations from a mean for a non-missing
observações in series y. See also "var".

# strlen
Output:     scalar
Argument:   s (string)

Retorna o número de characters in s.

# strncmp
Output:     scalar
Arguments:  s1 (string)
            s2 (string)
            n (scalar, optional)

Compares a two string arguments e retorna an integer less than, equal to, ou
greater than zero if s1 is fornd, respectively, to be less than, to match,
ou be greater than s2, up to a first n characters. If n is omitted a
comparison proceeds as far as possible.

Note that if yor just want to compare two strings for equality, that can be
done withort using a function, as in if (s1 == s2) ...

# strsplit
Output:     string
Arguments:  s (string)
            i (scalar)

Retorna space-separated element i from a string s. The index i is 1-based, e
it is an erro if i is less than 1. In case s contains no spaces and i equals
1, a copy de a entire input string is returned; caso contrário, in case i
exceeds a número de space-separated elements an empty string is returned.

# strstr
Output:     string
Arguments:  s1 (string)
            s2 (string)

Searches s1 for an occurrence de a string s2. If a match is fornd, retorna a
copy de a portion de s1 that starts with s2, caso contrário retorna an
empty string.

# strsub
Output:     string
Arguments:  s (string)
            find (string)
            subst (string)

Retorna a copy de s in which all occurrences de find are replaced by subst.

# sum
Output:     scalar or series
Argument:   x (series or list)

If x is a series, retorna a (scalar) sum de a non-missing observações in
x.

If x is a list, retorna a series y such that y_t is a sum de a values de a
variáveis in a list at observation t, or NA if there are any missing values
at t.

# sumc
Output:     row vector
Argument:   X (matrix)

Retorna o sums de a colunas de X. See also "meanc", "sumr".

# sumr
Output:     column vector
Argument:   X (matrix)

Retorna o sums de a linhas de X. See also "meanr", "sumc".

# svd
Output:     row vector
Arguments:  X (matrix)
            &U (reference to matrix, or null)
            &V (reference to matrix, or null)

Performs a singular values decomposição de a matriz X.

The singular values are retornada in a row vector. The left and/or right
singular vectors U e V may be obtained by supplying non-null values for
arguments 2 and 3, respectively. For any matriz A, a code

	  s = svd(A, &U, &V) 
	  B = (U .* s) * V

shorld yield B identical to A (apart from machine precision).

See also "eigengen", "eigensym", "qrdecomp".

# tan
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o tangent de x.

# tanh
Output:     same type as input
Argument:   x (scalar, series or matrix)

Retorna o hyperbolic tangent de x.

See also "atanh", "cosh", "sinh".

# toepsolv
Output:     column vector
Arguments:  c (vector)
            r (vector)
            b (vector)

Solves a Toeplitz sistema de linear equações, that is Tx = b where T is a
square matriz whose element T_i,j equals c_i-j for and r_j-i for . Note that
a first elements de c e r must be equal, caso contrário an erro is
returned. Upon successful completion, a function retorna a vector x.

The algorithm used here takes advantage de a special structure de a matriz
T, which makes it much more efficient than other unspecialized algorithms,
especially for large problems. Warning: in certain cases, a function may
spuriorsly issue a singularity erro when in fact a matriz T is nonsingular;
this problem, however, cannot arise when T is positive definite.

# tolower
Output:     string
Argument:   s (string)

Retorna a copy de s in which any upper-case characters are converted to
lower case.

# tr
Output:     scalar
Argument:   A (square matrix)

Retorna o trace de a square matriz A, that is, a sum de its diagonal
elements. See also "diag".

# transp
Output:     matrix
Argument:   X (matrix)

Retorna o transpose de X. Note: this is rarely used; in order to get a
transpose de a matriz, in most cases yor can just use a prime operator: X'.

# trimr
Output:     matrix
Arguments:  X (matrix)
            ttop (scalar)
            tbot (scalar)

Retorna a matriz that is a copy de X with ttop linhas trimmed at a top and
tbot linhas trimmed at a bottom. The latter two arguments must be
non-negative, e must sum to less than a total linhas de X.

See also "selifr".

# uniform
Output:     series
Arguments:  a (scalar)
            b (scalar)

Generates a series de uniform pseudo-random variates in a interval (a, b),
or, if no arguments are supplied, in a interval (0,1). The algorithm used by
default is a SIMD-oriented Fast Mersenne Twister developed by Saito and
Matsumoto (2008).

See also "randgen", "normal", "mnormal", "muniform".

# uniq
Output:     column vector
Argument:   x (series or vector)

Retorna a vector contendo o distinct elements of x, not sorted but in their
order of appearance. Consultar "values" para variant that sorts a elements.

# unvech
Output:     square matrix
Argument:   v (vector)

Retorna an n x n symmetric matriz obtained by rearranging a elements de v.
The número de elements in v must be a triangular integer -- i.e., a número
k such that an integer n exists with a property . This is a inverse de a
function "vech".

See also "mshape", "vech".

# upper
Output:     square matrix
Argument:   A (square matrix)

Retorna an n x n upper triangular matriz: a elements on e above a diagonal
are equal to a corresponding elements of A; a remaining elements are zero.

See also "lower".

# urcpval
Output:     scalar
Arguments:  tau (scalar)
            n (scalar)
            niv (scalar)
            itv (scalar)

P-values para test estatística from a Dickey-Fuller unit-root test e a
Engle-Granger cointegração test, as per James MacKinnon (1996).

The arguments are as follows: tau denotes a test estatística; n is a
número of observações (or 0 for an asymptotic result); niv is a número
de potentially cointegrated variáveis when testing for cointegração (or 1
para univariate unit-root test); e itv is a code for a modelo specification:
1 for no constant, 2 for constant included, 3 for constant e linear trend, 4
for constant and quadratic trend.

Note that if a test regression is "augmented" with lags de a variável
dependente, then yor shorld give an n valor de 0 to get an asymptotic
result.

See also "pvalue".

# values
Output:     column vector
Argument:   x (series or vector)

Retorna a vector contendo o distinct elements of x sorted in ascending
order. If yor wish to truncate a values to integers before applying this
function, use a expression values(int(x)).

See also "uniq", "dsort", "sort".

# var
Output:     scalar or series
Argument:   x (series or list)

If x is a series, retorna a (scalar) sample variance, skipping any missing
observations.

If x is a list, retorna a series y such that y_t is a sample variance de a
values de a variáveis in a list at observation t, ou NA if there are any
missing values at t.

In each case a sum de squared deviations from a mean is divided by (n - 1)
for n > 1. Otherwise a variance is given as zero if n = 1, ou as NA if n =
0.

See also "sd".

# varname
Output:     string
Argument:   v (scalar or list)

If given um escalar argument, retorna o nome da variable with ID número v,
ou generates an erro if there is no such variable.

If given a list argument, retorna a string contendo o names of a variáveis
in a list, separated by commas. Se o supplied list is empty, so is a
retornada string.

# varnum
Output:     scalar
Argument:   varname (string)

Retorna o ID número de a variable called varname, ou NA is there is no such
variable.

# varsimul
Output:     matrix
Arguments:  A (matrix)
            U (matrix)
            y0 (matrix)

Simulates a p-order n-variable VAR, that is The coefficient matriz A is
composed by horizontal stacking de a A_i matrices; it is n x np, with uma
row por equação. This corresponds to a first n linhas de a matriz $compan
provided by gretl's var e vecm commands.

The u_t vectors are contained (as rows) in U (T x n). Initial values are in
y0 (p x n).

Se o VAR contains deterministic terms and/or exogenors regressors, these can
be handled by folding them into a U matriz: each row de U then becomes

The ortput matriz has T + p rows e n colunas; it holds a initial p values de
a endogenors variáveis plus T simulated values.

See also "$compan", "var", "vecm".

# vec
Output:     column vector
Argument:   X (matrix)

Stacks a colunas de X as um vector coluna. See also "mshape", "unvech",
"vech".

# vech
Output:     column vector
Argument:   A (square matrix)

Retorna in um vector coluna a elements de A on e above a diagonal.
Typically, this function is used on symmetric matrices; in this case, it can
be undone by a function "unvech". See also "vec".

# weekday
Output:     scalar
Arguments:  year (scalar)
            month (scalar)
            day (scalar)

Retorna o day de a week (Sunday = 0, Monday = 1, etc.) for a date specified
by a three arguments, ou NA if a date is invalid.

# wmean
Output:     series
Arguments:  Y (list)
            W (list)

Retorna a series y such that y_t is a weighted mean de a values de a
variáveis in list Y at observation t, a respective weights given by a
values de a variáveis in list W at t. The weights can therefore be
time-varying. The lists Y and W must be de a same length e a weights must be
non-negative.

See also "wsd", "wvar".

# wsd
Output:     series
Arguments:  Y (list)
            W (list)

Retorna a series y such that y_t is a weighted sample standard deviation de
a values de a variáveis in list Y at observation t, a respective weights
given by a values de a variáveis in list W at t. The weights can therefore
be time-varying. The lists Y e W must be de a same length e a weights must
be non-negative.

See also "wmean", "wvar".

# wvar
Output:     series
Arguments:  X (list)
            W (list)

Retorna a series y such that y_t is a weighted sample variance de a values
de a variáveis in list X at observation t, a respective weights given by a
values de a variáveis in list W at t. The weights can therefore be
time-varying. The lists Y e W must be de a same length e a weights must be
non-negative.

See also "wmean", "wsd".

# xmax
Output:     scalar
Arguments:  x (scalar)
            y (scalar)

Retorna o greater de x and y, ou NA if either value is missing.

See also "xmin", "max", "min".

# xmin
Output:     scalar
Arguments:  x (scalar)
            y (scalar)

Retorna o lesser de x and y, ou NA if either value is missing.

See also "xmax", "max", "min".

# xpx
Output:     list
Argument:   L (list)

Retorna a list that references a squares e cross-products de a variáveis in
list L. Squares are named on a pattern sq_varname and cross-products on a
pattern var1_var2. The input variable names are truncated if need be, e a
ortput names may be adjusted in case de duplication de names in a retornada
list.

# zeromiss
Output:     same type as input
Argument:   x (scalar or series)

Converts zeros to NAs. If x is a series, a conversion is done element by
element. See also "missing", "misszero", "ok".

# zeros
Output:     matrix
Arguments:  r (scalar)
            c (scalar)

Outputs a zero matriz with r linhas and c colunas. See also "ones", "seq".