A time prediction model for residents consumption level based on ARIMA and PCA

Abstract

It is necessary but difficult to make a large number of observations on multiple variables reflecting the residents consumption level and collect a large amount of data for analysis to search for the rules. In this paper, a time prediction model for residents consumption level based on ARIMA and principal component analysis is proposed to solve this problem. Principal component analysis is firstly used to effectively reduce the number of indicators reflecting the residents consumption level. Combined with the ARIMA model, the residents consumption level is predicted. The results reflect the trend of residents consumption level towards the need for enjoyment and development materials on the basis of obtaining basic survival data.