基于非参数核估计方法的均值-VaR模型

发布时间：2018-02-10 01:41

本文关键词： 投资组合在险价值非参数核估计　出处：《中国管理科学》2017年05期 　论文类型：期刊论文

【摘要】：本文运用非参数核估计方法对资产组合的在险价值(Value at Risk,VaR)进行估计,得到VaR的非参数核估计公式,并基于VaR的非参数核估计公式建立投资组合选择模型。理论上该模型的目标函数具有良好的光滑性,便于优化问题求解。Monte Carlo模拟结果表明该模型具有大样本性质,估计误差会随着样本容量的增大而下降,且该模型在非对称和厚尾分布下的表现优于当前文献中常用的经验分布法和Cornish-Fisher展开法。基于我国上证50指数及其成份股实际数据的实证结果说明该模型是有效的。
[Abstract]:In this paper, the nonparametric kernel estimation method is used to estimate the value at risk of a portfolio of assets, and the nonparametric kernel estimation formula of VaR is obtained. A portfolio selection model is established based on VaR's nonparametric kernel estimation formula. Theoretically, the objective function of the model has good smoothness, which is easy to solve. Monte Carlo simulation results show that the model has a large sample property. The estimation error decreases as the size of the sample increases. The performance of the model under asymmetric and thick tail distribution is better than that of empirical distribution method and Cornish-Fisher expansion method commonly used in current literature. The empirical results based on the 50 index of Shanghai Stock Exchange and its constituent stocks show that the model is effective.
【作者单位】：广东财经大学金融学院;中山大学管理学院;
【基金】：国家自然科学基金资助项目(71231008,71603058,71573056) 教育部人文社会科学研究项目(16YJC790033) 广东省自然科学基金项目(2016A030313656,2015A030313629,2014A030310305) 广东省哲学社会科学规划项目(GD15YYJ06,GD15XYJ03) 广州市哲学社会科学规划项目(15Q20) 广州市社会科学界联合会2016年“羊城青年学人”研究项目(16QNXR08)
【分类号】：F224;F830.9

【相似文献】