Fundamental, Sentiment and Technical Analysis for Algorithmic Trading Using Novel Genetic Programming Algorithms

This thesis explores genetic programming (GP) applications in algorithmic trading, addressing significant advancements in the field. Investors typically rely on fundamental analysis (FA) or technical analysis (TA) indicators, with sentiment analysis (SA) gaining recent attention. Consequently, algorithms have become the primary method for developing pre-programmed trading strategies, leading to substantial financial benefits. While each analysis type has been studied individually, their combined exploration remains limited. Our motivation is to assess if integrating FA, SA, and TA indicators can improve financial profitability. Therefore, we propose the use of novel GP algorithms for the combination of the three analysis types, along with the use of a novel fitness function, and a novel GP operator that encourages active trading by injecting trees into the GP population that perform a high number of trades while achieving high profitability at low risk. To evaluate our GP variants’ performance, we conduct experiments on stocks of 42 international companies, comparing the novel algorithm with the GP variants introduced in the same chapter. Moreover, we compare the proposed GP algorithm against four machine learning benchmarks and a financial trading strategy. The evaluation employs three financial metrics: Sharpe ratio, rate of return, and risk. Results consistently show that the proposed GP algorithms in each chapter enhance the financial performance of trading strategies, surpassing the benchmarks.