期刊论文

Dai, Zhifeng

英文

International Review of Economics & Finance

1059-0560

2021

71

127-142

10.1016/j.iref.2020.09.006

This work was supported by the National Natural Science Foundation of China granted [ 71771030 , 11301041 , 71671018 ], Scientific Research Fund of Hunan Provincial Education Department [grant number 19A007 ], Hunan Province Graduate Research and Innovation Project [ CX2019702 ].

本校为第一且通讯机构

We find that combining de-noising stock returns by wavelet transform with new proposed technical indicators can significantly improve the accuracy of stock returns forecasts, in which the new technical indicators can directly reflect the trend of stock returns series. Empirical results indicate the stock returns forecasts generated by new technical indicators are statistically and economically significant both in-sample and out-of-sample prediction performance. And when multivariate information is used to predict stock returns, its predictabili...