基于模糊回归分析的投资组合选择模型  

Portfolio selection model based on fuzzy regression analysis

作　　者：柏林[1] 房勇[1] 

机构地区：[1]中国科学院数学与系统科学研究院,北京100190

出　　处：《系统工程理论与实践》2015年第7期1770-1776,共7页Systems Engineering-Theory & Practice

基　　金：国家自然科学基金(71271201;71431008)

摘　　要：近年来,在存在模糊性的金融市场中如何进行有效的投资组合管理吸引了学者们的关注,本文利用模糊线性回归对不同市场上长度不一致的股票数据进行了刻画和分析,并在改进的收益和协方差矩阵基础上构建了投资组合选择模型.算例结果表明,在股票数据长度不一致时,基于模糊回归分析的投资组合选择模型比截断数据的投资组合模型以及基于普通最小二乘回归的投资组合模型有更好的表现.In recent years, how to achieve an efficient portfolio in the financial market with fuzziness has attracted attention of scholars. In this paper, we use fuzzy linear regression to analyze series with different lengths, and build a portfolio selection model based on the improved expected return and covariance matrix. Numerical results show that for the series with different lengths, the portfolio selection model based on fuzzy regression performs better than both of the model based on truncated data and the model based on ordinary least squares regression.

关 键 词：模糊线性回归 投资组合选择 三角模糊数 长短数据 

分 类 号：F830[经济管理—金融学]