Modelling and Forecasting Stock Prices of Three Leading Companies in Sri Lanka Using Time Series Analysis

Abstract

In this research work, we tend to select the best time series model to forecast the stock prices in three main companies in Sri Lanka. We select the stock prices of Sampath Bank, Commercial Bank and John Keels Holdings. In this work, we use Autoregressive (AR), Moving average(MA) and Autoregressive Integrated Moving average(ARIMA) models. We compare the performances of these thee time series models in terms of the predictions and mean square errors. In simulation study, we predict the stock prices for ten days and also calculate the mean square error of the predictions. Autoregressive-Moving average (ARMA) model was found as the best time series model by comparing the mean square error.