An empirical study on stock price based on ARIMA mode

Abstract

The model in time series analysis are widely used in the field of economy, it can describe the historical data change over time, and analyze data to make predictions. The ARIMA is one of the important models in time series models. This paper summarizes several smooth data processing methods. Using the A-Information Criterion and the principle of the precision of the model, it gives the step and method to build ARIMA model. The ARIMA model is used to analyze the data of China Merchants Bank(600036) shares at the opening price(2013/01/04-2013/10/18) and to predict the next five days(2013/10/21-2013/10/25) stock opening price data. In contrast with the actual data, the model prediction error is smaller, indicating that ARIMA model is very suitable for short-term forecasts.