Algorithmic sign prediction and covariate selection across eleven international stock markets

作者：

Highlights：

• The results indicate that there may be some predictability in the stock markets.

• Some of the best trading strategies produce excess profits of some 3% annually.

• The efficient market hypothesis (EMH) cannot be completely rejected.

• Predicting large downturns may be easier than predicting profit or loss.

摘要

•The results indicate that there may be some predictability in the stock markets.•Some of the best trading strategies produce excess profits of some 3% annually.•The efficient market hypothesis (EMH) cannot be completely rejected.•Predicting large downturns may be easier than predicting profit or loss.

论文关键词：Stock market indices,S&P 500,Sign prediction,Efficient-market hypothesis,Regularized regression,Similarity-based classification

论文评审过程：Received 4 April 2018, Revised 12 June 2018, Accepted 29 July 2018, Available online 30 July 2018, Version of Record 15 August 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.07.061