Small sample properties of GARCH estimates and persistence

Soosung Hwang, Pedro L. Valls Pereira

European Journal of Finance · 2006 · 126 citations · 22 references

Concepts

TL;DR

The study proposes that at least 250 observations are needed for ARCH(1) models and 500 for GARCH(1,1) models to mitigate bias and convergence errors. The authors introduce a simple metric that quantifies how much GARCH conditional volatility accounts for squared returns. ML estimates of GARCH(1,1) are significantly negatively biased in small samples, convergence is often impossible under Bollerslev’s constraints, and the metric shows that conditional volatility explains little of squared returns when the ARCH parameter is small.

Abstract

Abstract It is shown that the ML estimates of the popular GARCH(1,1) model are significantly negatively biased in small samples and that in many cases converged estimates are not possible with Bollerslev's non-negativity conditions. Results also indicate that a high level of persistence in GARCH(1,1) models obtained using a large number of observations has autocorrelations lower than these ML estimates suggest in small samples. Considering the size of biases and convergence errors, it is proposed that at least 250 observations are needed for ARCH(1) models and 500 observations for GARCH(1,1) models. A simple measure of how much GARCH conditional volatility explains squared returns is proposed. The measure indicates that for a typical index return volatility whose ARCH parameter is very small, the conditional volatility hardly explains squared returns.

References

22