|
Brief communications
On a statistical criterion for the heterogeneity of second-order moments
S. G. Khaliullin N.I. Lobachevsky Institute of Mathematics and Mechanics, Kazan Federal University, 18 Kremlyovskaya str., Kazan, 420008 Russia
Abstract:
In this paper we consider the problem of heteroscedasticity in the theory of time series. The presence of heteroscedasticity (heterogeneity in second-order moments) in various data leads to known errors in statistical conclusions if it is not possible to notice this problem in time. This problem is most often encountered in the tasks of verifying the adequacy of a particular model in regression analysis or time series analysis. If the model is adequate, the residuals should be homoscedastic.
When studying the financial market, it is quite common to find heterogeneity of second-order moments in some financial indices, such as logarithmic profit in stock prices.
The paper considers a criterion for testing the hypothesis of homoscedasticity in statistical data.
Keywords:
heteroscedasticity, conditionally Gaussian models, volatility.
Received: 02.06.2022 Revised: 02.06.2022 Accepted: 29.06.2022
Citation:
S. G. Khaliullin, “On a statistical criterion for the heterogeneity of second-order moments”, Izv. Vyssh. Uchebn. Zaved. Mat., 2022, no. 8, 93–96; Russian Math. (Iz. VUZ), 66:8 (2022), 76–78
Linking options:
https://www.mathnet.ru/eng/ivm9806 https://www.mathnet.ru/eng/ivm/y2022/i8/p93
|
|