GARCH-M模型参数的经验似然估计

Empirical Likelihood Parametric Estimation for GARCH-M Models

本文研究了GARCH-M模型参数的统计推断问题.我们通过经验似然方法构造了模型参数的检验统计量,并证明了该统计量渐近眼从x^2分布.基于此统计量,我们应用截面似然方法进一步构造了市场相对风险厌恶δ的检验统计量,同时证明了该统计量仍然服从x^2分布.最后进行了数值模拟,模拟结果表明经验似然方法表现良好.

We study inference on parameters of GARCH-M models in this paper. The empirical likelihood method is used to construct test statistics. Under mild conditions, statistics are shown to have asymptotic X2 distributions. Based on theses statistics, test statistics for the relative risk aversion δ of a market are constructed by profile likelihood idea and are shown to be asymptotically distributed as X2 distribution. Simulations show that the proposed empirical likelihood statistics behave well.