Premium
Alternative estimators for factor garch models—A monte carlo comparison
Author(s) -
Lin WenLing
Publication year - 1992
Publication title -
journal of applied econometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.878
H-Index - 99
eISSN - 1099-1255
pISSN - 0883-7252
DOI - 10.1002/jae.3950070304
Subject(s) - estimator , monte carlo method , statistics , univariate , autoregressive conditional heteroskedasticity , mathematics , bivariate analysis , econometrics , confidence interval , multivariate statistics , volatility (finance)
This paper proposes four estimators for factor GARCH models: two‐stage univariate GARCH (2SUE), two‐stage quasi‐maximum likelihood (2SML), quasi‐maximum likelihood with known factor weights (RMLE), quasi‐maximum likelihood with unknown factor weights (MLE). A Monte‐Carlo study is designed for bivariate one‐factor GARCH models to examine the finite sample properties. Results are presented for biases, ratios of standard errors to standard deviations, ratios of variances, coverage of confidence intervals, effects of misspecified factor weights, and finite sample properties of the 2SUE for factor GARCH‐in‐mean models.