Premium
ON THE STABILITY OF A HETEROSCEDASTIC PROCESS
Author(s) -
Weis Andrew A.
Publication year - 1986
Publication title -
journal of time series analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.576
H-Index - 54
eISSN - 1467-9892
pISSN - 0143-9782
DOI - 10.1111/j.1467-9892.1986.tb00497.x
Subject(s) - heteroscedasticity , mathematics , econometrics , stability (learning theory) , variance (accounting) , series (stratigraphy) , conditional variance , regression analysis , variable (mathematics) , statistics , autoregressive conditional heteroskedasticity , economics , computer science , biology , volatility (finance) , paleontology , mathematical analysis , accounting , machine learning
Abstract. In this paper we derive the stability conditions in a time series regression model with a particular form of conditional heteroscedasticity. The variables affecting the variance include lagged errors, lagged dependent variables and a forecast variable. In addition to the usual stability conditions in a dynamic model, a condition on the parameters of the heteroscedasticity equation is also required.