Forecasting future trends: a comprehensive analysis of korea composite stock price index using advanced predictive models

Stock markets play a vital role in the economy of any nation, providing capital to businesses and avenues for investors to share in the profits of corporations. Stock price prediction is a complex task due to the various factors affecting market fluctuations. The potential for substantial financial gain is an important motivator for accurate forecasting in the stock market. This study proposes a hybrid model for forecasting that combines Decision Trees (DTs), Wavelet Transform (WT), and Battle Royale Optimizer (BRO). The model is further improved by the inclusion of decomposition and optimization techniques with robust data preparation. The analysis, based on the Korea Composite Stock Price Index (KOSPI) from January 1st, 2015, to July 25th, 2023, reveals that the WT-BRO-DT model significantly outperforms other methods by achieving a coefficient of determination value of 0.992. This research addresses the existing gaps in stock market forecasting by integrating advanced machine learning techniques that improve the accuracy of the predictions. While previous models have shown limitations in handling complex data and achieving high accuracy, this work offers a more reliable solution to help investors and analysts make better-informed decisions, reduce risk, and optimize returns. These findings have practical implications for financial decision-making and contribute to the improvement in the accuracy of forecasts of stock markets.