[R-SIG-Finance] GARCH models that use range data

[Vivek Rao]

Packages such as rugarch fit GARCH models to returns. Usually one fits the model to close-to-close (CC) returns. Is there an R package or R code that uses OHLC data to predict CC return volatility within a GARCH framework? The basic GARCH(1,1) model

s(t+1)^2 = w + a*r(t)^2 + b*s(t)^2

where s(t) is the conditional standard deviation and r(t) is the return could be augmented by a term depending on the high-low range hl(t) = log(high(t)/low(t)) to give

s(t+1)^2 = w + a*r(t)^2 + b*s(t)^2 + c*hl(t)^2

Vivek Rao