uGARCHdistribution-class: class: Univariate GARCH Parameter Distribution Class
In rgarch: Flexible GARCH modelling in R

Description Objects from the Class Extends Methods Note Author(s) See Also Examples

Class for the univariate GARCH Parameter Distribution.

A virtual Class: No objects may be created from it.

Class "GARCHdistribution", directly. Class "rGARCH", by class "GARCHdistribution", distance 2.

as.data.frame: signature(x = "uGARCHdistribution"): Extracts various values from object (see note).
plot: signature(x = "uGARCHdistribution", y = "missing"): Parameter Distribution Plots.
show: signature(object = "uGARCHdistribution"): Parameter Distribution Summary.

The as.data.frame function takes optionally 2 additional arguments, namely window which indicates the particular distribution window number for which data is required (is usually just 1 unless the recursive option was used), and which indicating the type of data required. Valid values for the latter are “rmse” for the root mean squared error between simulation fit and actual parameters, “stats” for various statistics computed for the simulations such as log likelihood, persistence, unconditional variance and mean, “coef” for the estimated coefficients (i.e. the parameter distribution and is the default choice), and “coefse” for the estimated robust standard errors of the coefficients (i.e. the parameter standard error distribution).
The plot method offers 4 plot types, namely, Parameter Density Plots (take window as additional argument), Bivariate Plots (take window as additional argument), Stats and RMSE (only when recursive option used) Plots. The standard option for which is used, allowing for a numeric arguments to one of the four plot types else interactive choice via “ask”.

Alexios Ghalanos

Classes uGARCHforecast, uGARCHfit and uGARCHspec.

## Not run: 
data(sp500ret)
spec = ugarchspec(variance.model=list(model="gjrGARCH", garchOrder=c(1,1)), 
		mean.model=list(armaOrder=c(1,1), arfima=FALSE, include.mean=TRUE, 
		garchInMean = FALSE, inMeanType = 1), distribution.model="std")
		
fit = ugarchfit(data=sp500ret[, 1, drop = FALSE], out.sample = 0, 
				spec = spec, solver = "solnp")
	
dist = ugarchdistribution(fit, n.sim = 2000, n.start = 50, m.sim = 5)

## End(Not run)