The stock market plays an integral role in the economic performance of a country, serving as a means to assess financial systems. Accurately predicting stock price movement is a complicated task because it takes into account multiple positive and negative aspects that are social, political, and economic, such as earnings announcements, market trends, and fluctuations in economies all over the world. This research presents a new and effective hybrid approach to stock price prediction. Using the Random Forest (RF) machine learning technique in combination with the Aquila Optimizer (AO) presents better predictive performances. The methodology used stock market data from the Shanghai Stock Exchange (SSE) from 2015 to 2023 and compares the proposed AO-RF model using other RF models using different optimization techniques, such as the Grasshopper Optimization Algorithm and biologically inspired optimization, and discusses challenges. Various tests were performed on the AO-RF model and showed that the AO-RF model outperformed every model regarding predictive accuracy and reliability. The AO-RF model is capable of capturing complex, dynamic, and uncertain trends in stock and capital data, and can serve as a valuable empirical tool for a wide range of investors. More reliable forecasts can improve investment thinking, which in turn will ultimately help investors make informed financial decisions and measure volatility levels in predictions.
Afandiyeva Hajar (Sun,) studied this question.