Forecasting stock price in healthcare sector by using geometric Brownian Motion model

This study addresses the challenge of selecting suitable stocks amid the daily fluctuations and unstable conditions of financial markets. It focuses on the critical need of forecasting tools to predict the future prices that able to assist investor in investment and avoid financial losses. Unlike many studies that focus on the long-term forecasting methods, this study uses Geometric Brownian Motion (GBM) model to forecast short-term prices-based investor desire. The model’s efficacy is tested using a case study of 10 healthcare stocks, with forecasts generated for two-week and four-week periods. The methodology involves modelling stock price using GBM model, which core component included the rate of return, drift and volatility. The accuracy of the forecasts is validated using the Mean Absolute Percentage Error (MAPE) and percentage increment. The performance of forecasted prices is evaluated against the FBMKLCI market index. Additional financial ratio including Sharpe’s and Treynor’s indices along with t-test to ascertain risk-adjusted returns and statistical significance. The finding shows that two-weeks forecast period yield the smallest MAPE, indicating that the GBM model provides a highly accurate forecast for short-term investment. A comparison of forecasted ranking with the actual price ranking reveals an 88% correlation, further supporting the model’s reliability. Additionally, most of the selected stocks demonstrate a significant correlation with the FBMKLCI. This study contributes to the academic discourse by empirically demonstrating the effectiveness of the GBM model in short-term forecasting in an emerging market, offering valuable insights for investors and financial analysts.