Спатиальная модель для оценки эффектов перетекания волатильности на рынке нефти и газа
The article is devoted to the estimation of volatility spillovers occurred on the oil and gas market taking into account cross-sectional dependence. In this paper we explore data on daily stock returns of 67 companies from oil and gas sector from 13 countries. The volatility spillovers are estimated via spatial specification of the BEKK model.
By means of Vuong test we compare the explaining power of spatial BEKK and non-spatial GO-GARCH и ADCC, Diebold-Mariano and Hansen-Lunde-Nason tests are used for evaluating the predictive ability of the models under consideration. Vuong test reveals equal explaining ability for the three models on any reasonable significance level.
Within the out-of-sample comparison, the tests do not provide clear evidence of the significant superiority of the spatial specification over the other models.