Analysis and Strategy Research on the Development Trend of Pet Industry Based on Multi Model Prediction

This study uses comprehensive data analysis and modeling methods to predict the future trends of the pet industry, offering data support and strategic recommendations for decision-making. By collecting data on the Chinese pet market and incorporating market data from the U.S., Germany, France, and globally, it analyzes market size, food demand, trade, and policies using Lasso regression, GM $(1,1)$ grey prediction, ARIMA, metabolic GM $(1,1)$, tariff elasticity, and multiple regression models. Missing and outlier values were cleaned, and key factors were screened using Pearson correlation analysis. The Lasso model predicts China's market size to reach 3208.43 billion yuan in 2024, 3503.68 billion yuan in 2025, and 3813.45 billion yuan in $2026\left(R^{2}=0.97\right)$. The ARIMA model forecasts global pet food demand at $128.96$ billion, $134.32$ billion, and $139.68$ billion in the next three years, with less than 5 % error. The metabolic GM $(1,1)$ model predicts China's pet food export value will increase from $52.34 billion to $90.08$ billion between 2024 and 2026. A tariff elasticity model calculates the effects of tariff changes on trade and the pet food market, considering exchange rates and other factors. The study concludes with strategies to promote sustainable industry development and offers valuable insights for policy and corporate strategy. The multi-model approach shows high prediction accuracy and offers significant guidance for the pet industry's future development.