Animal Breeding Customer Demand Forecasting Model Based on Outlier Detection Algorithm

  • Geer Teng
Keywords: Rough Set, Discrete Point Detection Algorithm, Animal Breeding, Customer Demand

Abstract

With the gradual concentration of animal breeding in China and the industrial transfer caused by
economic development in some areas, the time and space barriers of animal breeding and consumption have
become more and more obvious, and the oversupply or shortage of regional animals has become more common.
In order to accurately predict the needs of animal breeding customers, this paper analyzes from the perspective
of cost, and uses the discrete point detection algorithm based on rough set theory to construct and analyze the
customer demand model. After that, this paper conducts an empirical study on customer needs. Finally, based on
the above research content, the following conclusions are drawn: Through empirical research, it is proved that
the nonlinear network model is combined with the prediction result of the gray system GM (1, 1) model to
establish nonlinearity. The combined forecasting model is feasible to predict the forecast value of the customer
demand in China, and its prediction accuracy is higher than the above single forecasting models. The price of
live pigs is characterized by significant cyclical fluctuations. From the elastic analysis, the supply and demand
of pigs lack price elasticity, but the quantity and price are flexible for the supply of pigs. The demand for live
pigs lacks income elasticity and cross-price elasticity. The stability of the live pig market requires the state to
regulate and control the production of stable pigs and the protection of breeding interests.

Published
2019-09-01