基于MCMC方法的投资组合选择
首发时间:2012-11-30
摘要:本文在马克威茨的投资组合模型的基础上,建立了均值半绝对价值离差投资组合选择模型。在模型估计方面,为了更有效地解释由参数的不确定性导致的估计风险,运用了层次贝叶斯方法和MCMC对均值半绝对价值离差模型进行了分析,并对已提出的不同方法的结果进行了比较和分析。实证分析表明,该新模型和完全层次贝叶斯估计方法使得资产组合年绩效提高了1.5%。此外,该求解方法还可应用到多种市场收益效用函数和模型中。
关键词: 贝叶斯 投资组合 半绝对价值离差 MCMC 模拟退火
For information in English, please click here
Portfolio Selection Based on MCMC Method
Abstract:This paper constructed the mean semi-absolute value deviation model based on the Markowitz's portfolio selection study. In the respect of model estimation, in order to explain the estimation risk caused by the parameter uncertainty more valid, we introduced the hierarchy Bayesian method and the MCMC method to analyze the mean semi-absolute value deviation model. By empirical study, the result show that the new portfolio selection model and the hierarchy Bayesian method contributes to the performance of the initial portfolio by improving 1.5% annally. The method can be used in different market utility function and model.
Keywords: Bayesian portfolio selection mean semi-absolute value deviation MCMC Simulated Annealing
论文图表:
引用
No.****
同行评议
共计0人参与
勘误表
基于MCMC方法的投资组合选择
评论
全部评论0/1000