Statistical Comparison ARIMA Order Performance In Stock Market

Abstract

Stock market forecasting is important for financial decision-making and risk management. Among the various time series models, the Autoregressive (p) Integrated (d) Moving Average (q) (ARIMA) model has been widely adopted for its simplicity and effectiveness in capturing temporal patterns. However, selecting appropriate ARIMA orders remains a crucial and challenging task, impacting the accuracy of predictions. This paper presents a comprehensive statistical comparison of ARIMA order performance in the context of stock market forecasting. We examine the impact of different ARIMA model orders. Our study utilizes historic New York Stock Exchange (NYSE) stock price data. Our findings shed light on the complex interplay between ARIMA parameters and predictive accuracy, offering valuable insights for robust financial forecasting.