基于LPRE和LASSO方法的股指追踪研究
    点此下载全文
引用本文:陈银钧,刘惠篮.基于LPRE和LASSO方法的股指追踪研究[J].经济数学,2020,(1):92-96
摘要点击次数: 392
全文下载次数: 0
作者单位
陈银钧,刘惠篮 (贵州大学 数学与统计学院, 贵州 贵阳 550025) 
中文摘要:将最小化乘积相对误差(LPRE)和最小绝对压缩选择算子(LASSO)方法应用到乘积回归模型,结合BIC信息准则实现股票指数的追踪,成功选取了26支对上证50指数影响较大的成分股,并比较了所提方法与线性模型下LASSO方法的表现,验证了所提方法的有效性.
中文关键词:乘积模型  LPRE方法  LASSO估计  BIC准则
 
Research on Stock Index Tracking Based on LPRE and LASSO Methods
Abstract:Because the stock price is non-negative data, the multiplicative regression model can be applied to such data. The LPRE and LASSO method are applied to the multiplicative regression model for tracking the stock index. And the BIC information criterion is used to select the penalty parameters. 26 constituent stocks of Shanghai 50 Index are successfully selected. The performance of LASSO method in linear model is compared with our method, which verifies the effectiveness of the proposed method.
keywords:multiplicative regression model  LPRE method  LASSO method  BIC criterion
查看全文   查看/发表评论   下载pdf阅读器