Stock returns prediction using kernel adaptive filtering within a stock market interdependence approach

作者：

Highlights：

• Stock returns are predicted using individual local models learned from other stocks.

• The proposal predicts stock returns using interconnections among stock markets.

• The aim is to use previous stock returns to predict the current day price change.

• The proposal outperforms autoregressive models and NNs in terms of Sharpe Ratio.

• The results suggest that the US market is more influenced by the European.

摘要

•Stock returns are predicted using individual local models learned from other stocks.•The proposal predicts stock returns using interconnections among stock markets.•The aim is to use previous stock returns to predict the current day price change.•The proposal outperforms autoregressive models and NNs in terms of Sharpe Ratio.•The results suggest that the US market is more influenced by the European.

论文关键词：Stock returns prediction,Sequential learning,Interdependence between markets,Kernel adaptive filtering.

论文评审过程：Received 22 November 2019, Revised 24 April 2020, Accepted 13 June 2020, Available online 26 June 2020, Version of Record 2 July 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.113668