Modeling of the relationship between corporate social responsibility and stock price with artificial neural network

Abstract

Corporate social responsibility (CSR) is an important issue for organization due to the increasing attention of the firm responsibility to their stakeholders. Most of the literature in this area attempt to use statistical method to investigate the relationship of CSR to various firm performance indexes such as stock price, cost of capital, and return on asset. However, most of the works have low adjusted R square value due to the complexity of problem which leads to validity of the conclusion derived from that model. This paper presents the application of artificial neural network (ANN), one of the machine learning tools capable of modeling complex and non-linear relationship between variables. In this work, data from Thai companies listed in Market for Alternative Investment (mai) were used to train ANN. The trained ANN achieved significantly high adjusted R square value comparing with the statistical method.