×

Estimating GARCH models: When to use what? (English) Zbl 1135.91405

Summary: The class of generalized autoregressive conditional heteroscedastic (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 quasi-likelihood estimator based on a Gaussian likelihood (GMLE) and the log-transform-based least absolutely deviations estimator (LADE) [L. Peng and Q. Yao, Biometrika 90, 967–975 (2003)]. A practically relevant question is when to use what. We provide in this paper a solution to this question. By interpreting the LADE as a version of the maximum quasilikelihood estimator under the likelihood derived from assuming hypothetically that the log-squared innovations obey a Laplace distribution, we outline a selection procedure based on some goodness-of-fit type statistics. The methods are illustrated with both simulated and real data sets. Although we deal with the estimation for GARCH models only, the basic idea may be applied to address the estimation procedure selection problem in a general regression setting.

MSC:

91B70 Stochastic models in economics
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] DOI: 10.1111/1467-9868.00391 · Zbl 1065.62094 · doi:10.1111/1467-9868.00391
[2] DOI: 10.3150/bj/1068128975 · Zbl 1064.62094 · doi:10.3150/bj/1068128975
[3] Chow Y. S., Probability Theory, 3. ed. (1997) · doi:10.1007/978-1-4612-1950-7
[4] Fan J., Nonlinear Time Series: Nonparametric and Parametric Methods (2003) · Zbl 1014.62103 · doi:10.1007/b97702
[5] DOI: 10.1017/S0266466600161018 · Zbl 0986.60030 · doi:10.1017/S0266466600161018
[6] DOI: 10.1017/S0266466601173056 · Zbl 1051.62074 · doi:10.1017/S0266466601173056
[7] DOI: 10.1111/1468-0262.00396 · Zbl 1136.62368 · doi:10.1111/1468-0262.00396
[8] DOI: 10.1016/S0378-3758(02)00488-3 · Zbl 1032.62084 · doi:10.1016/S0378-3758(02)00488-3
[9] DOI: 10.1093/biomet/83.4.875 · Zbl 0883.62004 · doi:10.1093/biomet/83.4.875
[10] DOI: 10.1016/0888-613X(94)90007-8 · Zbl 0794.68132 · doi:10.1016/0888-613X(94)90007-8
[11] DOI: 10.2307/2171862 · Zbl 0844.62080 · doi:10.2307/2171862
[12] DOI: 10.1214/009053605000000840 · Zbl 1091.62082 · doi:10.1214/009053605000000840
[13] DOI: 10.1093/biomet/90.4.967 · Zbl 1436.62439 · doi:10.1093/biomet/90.4.967
[14] Straumann D., Estimation in Conditionally Heteroscedastic Time Series Models (2005) · Zbl 1086.62103
[15] DOI: 10.1214/009053606000000803 · Zbl 1108.62094 · doi:10.1214/009053606000000803
[16] Takeuchi K., Mathematical Sciences 153 pp 12– (1976)
[17] Weiss A., Econometric Theory 2 pp 107– (1986)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.