Forecast of Chinese pet population and Global Pet Food Demand Based on RF-MLR-ARIMA and GB-XGBoost

Yining Gao

doi:10.54097/e43ph702

Authors

Yining Gao

DOI:

https://doi.org/10.54097/e43ph702

Keywords:

Pet Industry, Pet Population Prediction, Pet Food Demand, Integrated Prediction Model, Gradient Boosting Regression – XGBoost.

Abstract

With the global economic development, the acceleration of urbanization and the wide popularization of the "pet companionship" concept, the pet industry has grown from a traditional niche field into an emerging economic sector with both market potential and development challenges. Accurately analyzing the trends of core industry indicators has become a key prerequisite for supporting the sustainable development of the industry and the strategic decision-making of enterprises. This paper constructs a comprehensive prediction model of random forest - multiple linear regression - ARIMA, and concludes that the number of pet cats in China will continue to increase in the next three years, while the number of pet dogs will tend to stabilize. After the dimensionality reduction of the indicators, the gradient boosting regression - XGBoost model was adopted for analysis. The results show that the United States dominates the global demand for pet food, while the Chinese market demonstrates strong growth. This research can provide strong academic support for the trend analysis of the pet industry and the formulation of sustainable development strategies.

References

[1] Li, J., Wang, J., & Liu, Y. Forecasting China's pet market scale using an ARIMA-XGBoost hybrid model[J]. Journal of Agrotechnical Economics, 2023, (6): 132-140.

[2] Zhang, S. Y., Chen, M. H., & Li, X. T. Research on influencing factors and prediction of global pet food demand based on gradient boosting regression[J]. Journal of International Trade, 2022, (11): 98-109.

[3] Wang, Z. H., Liu, M., & Zhao, L. Dynamic prediction of China's pet population and analysis of driving factors[J]. Acta Veterinaria et Zootechnica Sinica, 2021, 52(8): 2456-2465.

[4] Liu, C., Li, M., & Wang, L. The impact of tariff policies on the import and export of pet products: An empirical analysis based on multiple linear regression[J]. Journal of International Trade, 2022, (8): 78-86.

[5] Chen, X., Zhao, L., & Sun, Y. Factor analysis and screening of core development indicators in the pet industry[J]. Commercial Research, 2023, (5): 34-41.

[6] Liu, Y., Li, M., & Zhang, H. Forecasting pet population in China using hybrid ARIMA-random forest model[J]. Journal of Animal and Veterinary Advances, 2023, 22(3): 189-201.

[7] Wang, H., Chen, J., & Lee, S. Demand prediction for global pet food industry based on XGBoost algorithm[J]. Pet Food Science and Technology, 2022, 15(2): 45-58.

[8] Smith, J. D., Johnson, L. K., & Davis, E. F. Multi-model ensemble for pet industry trend analysis: A case study of US and China[C]//2024 International Conference on Data Science and Intelligent Computing, 2024: 321-326.

[9] Kim, H. J., Park, J. H., & Lee, S. Y. Analysis of pet food export factors using machine learning models[J]. Journal of International Trade and Commerce, 2023, 19(3): 78-92.

[10] Brown, R. S., Wilson, T. G., & Anderson, K. L. Temporal forecasting of pet ownership trends with gradient boosting regression[J]. Animals, 2022, 12(15): 1987.

[11] Davis, M. E., Thompson, R. C., & Clark, S. E. Policy impacts on pet industry development: A panel data analysis[J]. Journal of Agricultural and Applied Economics, 2021, 53(1): 124-139.

[12] Wilson, K. D., Martin, J. S., & Harris, L. M. Time series forecasting of pet healthcare demand in North America[J]. Veterinary Record, 2023, 192(7): 245-251.