Leveraging Deep Learning for Predictive Analysis of S&P BSE SME IPO Index Close Price

Forecasts about stock markets represent a vital financial analysis problem due to their intensely unstable characteristic together with their non-linear price behavior. The model proposed in this study combines Long Short-Term Memory (LSTM) as well as Gated Recurrent Unit (GRU) to forecast closing prices of the S&P BSE SME IPO Index. The model processed historical data obtained from Investing.com, including open and high prices, as well as low prices and close prices. The evaluation results show that our model reaches 99.60% accuracy while revealing Mean Squared Error (MSE) of 29.42 and R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score value of 0.9991. The novel approach of this study combines LSTM with GRU layers as a whole to exploit their strengths that produce better results than standard models. Financial time series prediction now benefits from a deep learning-based methodology that provides efficient results which helps investors along with analysts when making decisions.