Value at risk predictions via density ratio modeling
Given m time series regression models,linear or not,with additive noise components,it is shown how to estimate semiparametrically the predictive probability distribution of one of the time series conditional on all the observed and covariate data at the time of prediction.This is done by a certain synergy argument,assuming that the distributions of the noise components associated with the regression models are tilted versions of a reference distribution.An application to the estimation of the value at risk (VaR) of financial returns is discussed.A simulation shows the present method competes well with both historical and GARCH VaR estimates under different distributional scenarios.
Semiparametric biased sampling empirical likelihood exponential tilt predictive distribution VaR.
Haiming Guo
Department of Finance Changzhou University
国内会议
上海
英文
703-716
2013-11-08（万方平台首次上网日期，不代表论文的发表时间）