R Studio

Heston Nandi Garch

## hngarchSim -
# Simulate a Heston Nandi Garch(1,1) Process:
# Symmetric Model - Parameters:

model = list(lambda = 4, omega = 8e-5, alpha = 6e-5,
             beta = 0.7, gamma = 0, rf = 0)
ts = hngarchSim(model = model, n = 500, n.start = 100)
par(mfrow = c(2, 1), cex = 0.75)
ts.plot(ts, col = "steelblue", main = "HN Garch Symmetric Model")
grid()

## hngarchFit -
# HN-GARCH log likelihood Parameter Estimation:
# To speed up, we start with the simulated model ...

mle = hngarchFit(model = model, x = ts, symmetric = TRUE)
mle
plot(mle,)

## summary.hngarch -
# HN-GARCH Diagnostic Analysis:

par(mfrow = c(3, 1), cex = 0.75)
summary(mle)

## hngarchStats -
# HN-GARCH Moments:

hngarchStats(mle$model)