Back to Search Start Over

Full versus quasi MLE for ARMA-GARCH models with infinitely divisible innovations.

Authors :
Goode, Jimmie
Kim, Young Shin
Fabozzi, Frank J.
Source :
Applied Economics; 2015, Vol. 47 Issue 48, p5147-5158, 12p, 5 Charts, 3 Graphs
Publication Year :
2015

Abstract

We compare the backtesting performance of ARMA-GARCH models with the most common types of infinitely divisible innovations, fit with both full maximum likelihood estimation (MLE) and quasi maximum likelihood estimation (QMLE). The innovation types considered are the Gaussian, Student’st,α-stable, classical tempered stable (CTS), normal tempered stable (NTS) and generalized hyperbolic (GH) distributions. In calm periods of decreasing volatility, MLE and QMLE produce near identical performance in forecasting value-at-risk (VaR) and conditional value-at-risk (CVaR). In more volatile periods, QMLE can actually produce superior performance for CTS, NTS andα-stable innovations. While thet-ARMA-GARCH model has the fewest number of VaR violations, rejections by the Kupeic and Berkowitz tests suggest excessively large forecasted losses. Theα-stable, CTS and NTS innovations compare favourably, with the latter two also allowing for option pricing under a single market model. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
00036846
Volume :
47
Issue :
48
Database :
Complementary Index
Journal :
Applied Economics
Publication Type :
Academic Journal
Accession number :
108580174
Full Text :
https://doi.org/10.1080/00036846.2015.1042203