A Research on Theft Crime Based on SARIMA Model and Improved X13-ARIMA-SEATS Model

Abstract

This thesis takes the theft crime data of a city as the research object and analyses the time pattern of theft crime, hoping to be able to provide the public security organs certain decision-making reference for preventing and reducing the theft crime. SARIMA model and improved X13-ARIMA-SEATS model were established based on the time series data. The model with higher precision accuracy is selected to analyse the time pattern of theft crime. Through modelling, we can respectively determine the SARIMA prediction model and the improved X13-ARIMA-SEATS model. After that, two models are used for in-sample prediction. By comparing the prediction accuracy of the two models, it can be found that the prediction accuracy of X13 - ARIMA-SEATS Model is higher. Therefore, X13-ARIMA-SEATS Model is chosen to analyse the optimal time model for theft crime. Through this model, we can analyse the time pattern of theft crime data.