glarmaSimModel: Create a glarma simulation model

In glarma: Generalized Linear Autoregressive Moving Average Models

Create a glarma simulation model

Description

Defines a glarma model for input to the function glarmaSim.

Usage

glarmaSimModel(X, beta, offset = NULL, type = "Poi", mBin = NULL,
               alpha = NULL, residType = "Pearson",
               phiLags = NULL, phi = NULL,
               thetaLags = NULL, theta = NULL)

Arguments

X	Matrix; the explanatory variables. A vector of ones should be added to the data matrix as the first column for the beta of the intercept.
beta	Numeric vector; values of the regression coefficients.
offset	Either NULL or a numeric vector of length equal to the number of cases.
type	Character; the count distribution. Possible values are "Poi" (Poisson), "Bin" (binomial) and "NegBin" (negative binomial). The default is the Poisson distribution.
mBin	Numeric vector of length equal to the forecast horizon; only for the binomial case, the number of trials for each time point.
alpha	Numeric; for the negative binomial case, the shape parameter for glm.nb
residType	Character; the type of residuals to be used. Possible values are "Pearson" and "Score", and for the binomial distribution "Identity" is also allowed. The default is to use Pearson residuals.
phiLags	Numeric vector; AR orders.
phi	Numeric vector; values for the corresponding AR orders.
thetaLags	Numeric vector; MA orders.
theta	Numeric vector; values for the corresponding MA orders.

Value

Creates a list object of type "glarmaSimModel" with the same list elements as in the function call.

Author(s)

"David J. Scott" <[e-mail]> and "William T.M. Dunsmuir" <[e-mail]>

See Also

See glarmaSim.

Examples

### Test glarmaSimModel
data("Polio")
y <- Polio[, 2]
X <- as.matrix(Polio[, 3:8])
glarmamod <- glarma(y, X, thetaLags = c(1,2,5), type = "Poi", method = "FS",
                    residuals = "Pearson", maxit = 100, grad = 1e-6)
summary(glarmamod)
PolioModel <- glarmaSimModel(X, beta = coef.glarma(glarmamod, type = "beta"),
                             phiLags = glarmamod$phiLags,
                             phi = rep(0.1, length(glarmamod$phiLags)),
                             thetaLags = glarmamod$thetaLags,
                             theta = rep(0.1, length(glarmamod$thetaLags)),
                             type = glarmamod$type,
                             residType = glarmamod$residType)
str(PolioModel)
coef.glarma(glarmamod, type = "ARMA")
PolioModel <- extractGlarmaSimModel(glarmamod)
str(PolioModel)

glarma documentation built on April 4, 2025, 12:32 a.m.