Investigating the Efficacy of ARIMA Models for Predicting Dow Jones Industrial Average Stock Prices

Abstract

The prediction of stock price volatility holds significant importance in the realms of economics and finance, offering substantial benefits to both investors and economists. This paper employs the Autoregressive Integrated Moving Average (ARIMA) model to forecast the stock prices of the Dow Jones Industrial Average (DJI), a key index in the financial market. The study utliises daily data of the Dow Jones Industrial Average (DJI) spanning from 1 April 2021 to 31 March 2023. Empirical evidence strongly suggests the effectiveness of ARIMA models in predicting DJI stock prices. Furthermore, the study’s findings reveal that the ARIMA model excels particularly in short-term forecasting, demonstrating favourable performance when compared to existing techniques for stock price prediction. Through the utilisation of the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) criteria, the study identifies ARIMA (1, 1, 0) as the optimal model for accurately forecasting the share price of DJI within the specified timeframe.

Keywords: ARIMA Model, Dow Jones Industrial Average (DJI), USA, Stationarity

View PDF