Estimation for Conditional Higher Moments Risk Based on Independent Component Analysis
Qifa Xu, Cuixia Jiang · 2006
With the development of the theory and practice in finance, higher moments risk has come to people's attention. Higher moments risk has the time-varying character, which is similar to the character of the second moment risk, i.e. variance risk. To reveal the relationship between higher moments risk in different markets or different financial assets at the same time, multivariate GARCHSK model is proposed in the paper. Estimation method for multivariate higher moments risk, which solved the problem of "dimension disaster" in multivariate GARCHSK modeling, is discussed in detail through independent component analysis. Finally, the method is applied to describe the conditional higher moments risk in international stock markets