logitxSim: Simulate from a dynamic logit-x model

In gets: General-to-Specific (GETS) Modelling and Indicator Saturation Methods

Simulate from a dynamic logit-x model

Description

Simulate from a dynamic Autoregressive (AR) logit model with covariates ('X'). This model is essentially a logit-version of the model of Kauppi and Saikkonen (2008).

Usage

logitxSim(n, intercept = 0, ar = NULL, xreg = NULL, verbose = FALSE, 
    as.zoo = TRUE)

dlogitxSim(n, ...)

Arguments

n	integer, the number of observations to generate
intercept	numeric, the value of the intercept in the logit specification
ar	NULL or a numeric vector with the autoregressive parameters
xreg	NULL or numeric vector with the values of the X-term
verbose	logical. If FALSE, then only the binary process (a vector) is returned. If TRUE, then a matrix with all the simulated information is returned (binary process, probabilities, etc.)
as.zoo	logical. If TRUE, then the returned object - a vector or matrix - will be of class zoo
...	arguments passed on to logitxSim

Details

No details, for the moment.

Value

A vector or matrix, depending on whether verbose is FALSE or TRUE, of class zoo, depending on whether as.zoo is TRUE or FALSE

Author(s)

Genaro Sucarrat, http://www.sucarrat.net/

References

Heikki Kauppi and Penti Saikkonen (2008): 'Predicting U.S. Recessions with Dynamic Binary Response Models'. The Review of Economic Statistics 90, pp. 777-791

See Also

logitx

Examples

##simulate from ar(1):
set.seed(123) #for reproducibility
y <- logitxSim(100, ar=0.3)

##more output (value, probability, logit):
set.seed(123) #for reproducibility
y <- logitxSim(100, ar=0.3, verbose=TRUE)

gets documentation built on Oct. 10, 2022, 1:06 a.m.