Research on the Pet Industry Based on Multiple Linear Regression Model

Abstract

The Chinese pet industry has a broad international market but also contends with complex international situations and a volatile market environment. Thus, scientifically analyzing and predicting its development trends is crucial. This paper examines recent data trends, using normality tests, correlation analysis, the entropy weight method, time series models, multiple linear regression models, and software like MATLAB, SPSS, and EXCEL to analyze and predict the industry's development and propose strategies. A visual line chart of pet market development was drawn based on data analysis to intuitively assess the situation. Normality and correlation analyses were done, yielding key results. The entropy weight method showed the comprehensive score increasing yearly, suggesting a rising market demand for pet food and related sectors. Leveraging collected data on global pet markets, an ARIMA time series model was built for future predictions. Multiple linear regression and multi-scenario analyses were also carried out, revealing an overall upward trend in the pet industry market.