Yasmin Akter Bipasha, a researcher affiliated with Westcliff University in the United States and Bangladesh University of Professionals, has contributed to the growing field of financial technology through her research paper titled “Stock Market Prediction Using LSTM.”
In this work, she addresses one of the most persistent challenges faced by investors in Bangladesh and around the world—the difficulty of accurately predicting stock prices in markets characterized by noise, uncertainty, and rapid change.
Beginning with a clear explanation of why traditional statistical models often fall short in capturing complex market behavior, her research gradually introduces Long Short-Term Memory (LSTM), a deep learning model specifically designed to handle time-series data and long-term dependencies.
Using real historical stock data from major global technology companies such as Apple, Amazon, Google, and Microsoft, the study demonstrates how LSTM models can more effectively capture temporal patterns in stock prices compared to conventional methods.
The significance of this research lies in its practical relevance: for developing markets like Bangladesh, where retail investors are increasingly entering the stock market with limited analytical tools, LSTM-based prediction models offer a data-driven approach to improve decision-making and risk management.
At the global level, the research contributes to ongoing efforts to integrate artificial intelligence into financial forecasting, supporting more informed investment strategies, reduced uncertainty, and improved market efficiency.
By combining methodological clarity with real-world applicability, Bipasha, also studied in Bangladesh’s University of Professionals, highlighted how machine learning can move stock market analysis beyond speculation toward more reliable, intelligent, and technology-driven financial forecasting systems with lasting impact for both local and international markets.
