| In the field of economic research,the vast majority of the sample data are not based on controlled experiments,but are produced during the normal operation of the economic system,so the change of an economic variable is usually not caused by a single cause variable,often the result of the joint action of multiple factors.What people often want to know is the quantitative relationship between the changes of a certain factor and the economic variables studied under other factors,and the researchers are often not satisfied with the approximate proportion of the two economic variables,but the more precise nonlinear relationship between the two.In this paper,a new nonlinear estimation method is proposed.Based on the minimum variance square sum criterion,a mathematical model for the quantitative study of the causality between two economic variables is established.The effects of the changes of other factors on the studied economic variables are eliminated as far as possible,while the selected factors are reserved as much as possible.The core of this paper is to give a mathematical model based on the sum of least squares of variance.The rationality of the minimum variance square sum criterion is explained from the characteristics of the economic meaning and the mathematical method,and the linear equations of the solution are derived.In this paper,an ideal sample is used to simulate the applicability of the theoretical results.On the one hand,the characteristics of the minimum variance square sum are explained.On the other hand,the validity of the minimum variance square sum criterion proposed in this paper is verified.Through theoretical research and simulation analysis,this paper aims to pr ovide new nonlinear parameter estimation methods for the researchers,and use the theory of least difference square sum to combine the actual economic pro blems and the linear equations derived from the model to provide effective stra tegic basis for the decision-makers. |
PDF Full Text Request