Effectiveness of Geometric Brownian Motion Method in Predicting Stock Prices: Evidence from India

Abstract

This research examines whether stock prices in the Indian stock markets follow a Geometric Brownian Motion (GBM). This study is keen on knowing if one can predict the simulated stock prices accurately against the actual stock prices. One-year, three-year, and five-year data of the historical stock prices of 50 stocks listed on the S&P BSE (Bombay Stock Exchange) Sensex 50 Index were employed as the base data to predict stock prices using the Monte Carlo simulation's GBM method. This study investigates whether there are statistically significant differences between the actual stock prices for three months and the simulated prices of the same period. This research has found that the GBM Monte Carlo simulation effectively predicts future stock prices for three months based on the historical data of stock prices of the past year. This study did not find significant differences between the actual and predicted stock prices when the simulation used the past one year's data. This research is original in the Indian context, as it situates the GBM method of Monte Carlo simulation in the premise of bounded rationality and efficient market hypothesis theories. There is thus the empirical evidence for bounded rationality and that the stock markets are not efficient.

Keywords: Monte Carlo simulation; stock prediction tools; efficient market hypothesis; bounded rationality; geometric brownian motion