The ability to accurately predict future events is crucial in the financial industry, as it allows for the determination of profit or loss. Researchers have dedicated techniques to forecasting in order to help money dealers achieve their profit goals, and stock prediction is a major issue on which investors rely when making investment decisions. For many different types of businesses, stock market data is notoriously unpredictable and erratic, and using the wrong model for time series data can lead to inaccurate forecasts. The research problem of finding a reliable and accurate method to predict Turkish stock market indicators is based on the fact that in order to get forecast models of stock market data that can faithfully depict reality and produce future forecasts, these models need to take into account all data factors, including linear and non-linear trends, various influences, and other data elements. Of Turkey’s financial markets, 83% are banking systems, measured in terms of asset size. This research provides a price forecasting analysis of the Borsa Istanbul Banks Index (representing the domestic banking sector) from Modified Gated Recurrent Unit (MGRU). The index is significant in the Turkish capital market. It is the Modified Jellyfish Search Optimization Algorithm (MJFSOA) that does the model fine-tuning for the MGRU. According to the results, the MGRU outperforms the competing models. Researchers at the technique selection phase of working with time series data, as well as investment businesses and managers making stock price movement forecasts, can benefit from the results.