Likelihood-Related Estimation Methods and Non-Gaussian GARCH Processes
نویسندگان
چکیده
This article discusses the finite distance properties of three likelihood-based estimation strategies for GARCH processes with non-Gaussian conditional distributions: (1) the maximum likelihood approach; (2) the Quasi Maximum Likelihood approach; (3) a multi-steps recursive estimation approach (REC). We first run a Monte Carlo test which shows that the recursive method may be the most relevant approach for estimation purposes. We then turn to a sample of SP500 returns. We confirm that the REC estimates are statistically dominating the parameters estimated by the two other competing methods. Regardless of the selected model, REC estimates deliver the more stable results.
منابع مشابه
On the efficiency of a semi-parametric GARCH model
Financial time series exhibit time-varying volatilities and non-Gaussian distributions. There has been considerable research on the GARCH models for dealing with these issues related to financial data. Since in practice the true error distribution is unknown, various quasi maximum likelihood methods based on different assumptions on the error distribution have been studied in the literature. Ho...
متن کاملThe effect of parameter estimation on Phase II control chart performance in monitoring financial GARCH processes with contaminated data
The application of control charts for monitoring financial processes has received a greater focus after recent global crisis. The Generelized AutoRegressive Conditional Heteroskedasticity (GARCH) time series model is widely applied for modelling financial processes. Therefore, traditional Shewhart control chart is developed to monitor GARCH processes. There are some difficulties in financial su...
متن کاملEstimating GARCH Models: When to Use What?
The class of GARCH models has proved particularly valuable in modelling time series with time varying volatility. These include financial data, which can be particularly heavy tailed. It is well understood now that the tail heaviness of the innovation distribution plays an important role in determining the relative performance of the two competing estimation methods, namely the maximum quasilik...
متن کاملProperties and Estimation of GARCH(1,1) Model
We study in depth the properties of the GARCH(1,1) model and the assumptions on the parameter space under which the process is stationary. In particular, we prove ergodicity and strong stationarity for the conditional variance (squared volatility) of the process. We show under which conditions higher order moments of the GARCH(1,1) process exist and conclude that GARCH processes are heavy-taile...
متن کاملRank-Based Estimation for GARCH Processes
We consider a rank-based technique for estimating GARCH model parameters, some of which are scale transformations of conventional GARCH parameters. The estimators are obtained by minimizing a rank-based residual dispersion function similar to the one given in Jaeckel (1972). They are useful for GARCH order selection and preliminary estimation. We give a limiting distribution for the rank estima...
متن کامل