All Rights Reserved
AccessEcon LLC 2006, 2008.
Powered by MinhViet JSC

 
Tucker S McElroy
 
''Stationary parameterization of GARCH processes''
( 2022, Vol. 42 No.4 )
 
 
We propose using the multivariate logistic transform to re-parameterize the Autoregressive Conditionally Heteroscedastic model such that the necessary stationarity constraints are automatically imposed, thereby allowing for unconstrained optimization when computing quasi-maximum likelihood estimates. A few simulations and a standard R data set of daily closing prices (Germany DAX) provide illustrations of the re-parameterization. We offer some numerical comparisons to available R packages (fgarch and rugarch), and comment on the potential advantages of the new technique.
 
 
Keywords: Conditional Heteroscedasticity, Stationarity, Time Series
JEL: C1 - Econometric and Statistical Methods: General
C5 - Econometric Modeling: General
 
Manuscript Received : Dec 28 2021 Manuscript Accepted : Dec 30 2022

  This abstract has been downloaded 28 times                The Full PDF of this paper has been downloaded 153784 times