Presenting an innovative methodology to effectively handle investment risk in financial markets

The stock market’s capital has grown swiftly, attracting an increasing number of investors. Given the high risk and high-profit margins associated with stock investments, investors want efficient schemes to assess the state of the market, forecast future trends, and choose profitable stocks. Researchers have constantly shown interest in stock projection, a basic topic at the crossroads of computer science and finance. The main objective of investing in financial markets is to maximize profits, considering the constantly changing conditions. Given this context, the research aims to create an accurate hybrid scheme for predicting stock prices by combining optimizers and extreme gradient boosting tactics. The optimization tactics that are followed in this exploration are particle swarm optimization (PSO), artificial bee colony (ABC), and ant lion optimization (ALO). The scheme chosen for this study, extreme gradient boosting, requires data as input to provide projections utilizing artificial intelligence-based schemes. The historical data used in this article include open prices, high, low, and close prices. These data were gathered to forecast the Korea Composite Stock Price Index stock market’s closing price from the beginning of January 2015 to the end of June 2023. It is noteworthy that this model when paired with the ALO, produced highly accurate and performant results, as evidenced by the regression coefficient of 0.9784. In the final analysis, the recommended model is a strong tool for investors in the financial market, offering a reliable means of treading through the complexities and uncertainties involved in stock price forecasting.