rmixnorm.ts: AR-type random variables from mixture of normals

In feng-li/flutils: Feng Li's utility functions

Description

Simulating AR type random variables from mixture of normals

Usage

rmixnorm.ts(n, means.ar.par.list, sigmas.list, weights, yinit = 0)

Arguments

n	a positive integer. Number of samples.
means.ar.par.list	parameters in AR(p) within each mixing component
sigmas.list	variance list
weights	weight in each list
yinit	initial values

Details

This function simulates random samples from a finite mixture of Gaussian distribution where the mean from each components are AR(p) process.

Value

vector of n follows a mixture distribution

Author(s)

Feng Li, Central University of Finance and Economics.

References

Li 2010 JSPI

Examples

n = 1000
means.ar.par.list = list(c(0, 0.8), c(0, 0.6, 0.3))
require("mvtnorm")
require("fGarch")
sigmas.spec <- list(garchSpec(model = list(alpha = c(0.05, 0.06)), cond.dist = "norm"),
                    garchSpec(model = list(alpha = c(0.05, 0.05)), cond.dist = "norm"))
sigmas.list <- lapply(lapply(sigmas.spec, garchSim, extended = TRUE, n = n),
                      function(x) x$sigma)
weights <- c(0.8, 0.2)
y = rmixnorm.ts(n = n, means.ar.par.list = means.ar.par.list, sigmas.list = sigmas.list,
                weights = weights)
plot(y, type = "l")

feng-li/flutils documentation built on Oct. 1, 2020, 9:09 p.m.