A Bayesian regularized artificial neural network for stock market forecasting

作者：

Highlights：

• A Bayesian regularization technique network optimization is proposed.

• Two major stocks were modeled to predict future prices.

• The model error is comparable to other advanced models.

• The Bayesian technique significantly reduces potential for overfitting.

摘要

•A Bayesian regularization technique network optimization is proposed.•Two major stocks were modeled to predict future prices.•The model error is comparable to other advanced models.•The Bayesian technique significantly reduces potential for overfitting.

论文关键词：Bayesian regularization,Neural network,Stock prediction,Overfitting

论文评审过程：Available online 19 April 2013.

论文官网地址：https://doi.org/10.1016/j.eswa.2013.04.013