The arima model for stock price prediction: studying the impact of news announcements through the use of agent-based modelling

Abstract

The relationship between new information and stock price movement is of great interest for investors. News announcements generally cause changes in stock prices, as they affect public perception of products and companies. An accurate predictive model allows us to take advantage of this unexpected movement. In this project, we use Agent-Based modelling to study the impact of news announcements on the ARIMA model’s accuracy. Overall, we conclude that a significant shock can cause long-term effects on the model’s precision.