Diverse Group Stock Portfolio Optimization Based on Investor Sentiment Index

Financial data analysis has always been an interesting issue and stock portfolio optimization is one of its most popular research topics. Recently, research on investor sentiment is also very popular in financial data analysis, and some scholars claim that it has impact on the stock market. In this paper, the proposed approach thus uses two investor sentiment indices and utilizes the grouping genetic algorithm to obtain a diverse group stock portfolio. Based on the fitness function defined in the previous approach, the adopted fitness function adds a new factor called risk of investor sentiment to evaluate the chromosomes for finding an appropriate diverse group stock portfolio. At last, experiments were conducted on a real stock dataset to verify the effectiveness of the proposed approach.