2019 年Metalearners for Estimating Heterogeneous Treatment Effects using Machine Learning的发表引发了世界因果推断理论的研究热情.目前,机器学习与因果推断论中的许多统计模型已被广泛应用.本文采用文献资料法、数理统计分析方法、案例研究法等研究方法,研究推演现代因果推断理论中较为知名的Uplift因果模型在体育中的应用场景,其中Uplift因果模型包括S-learner(单模型)、T-learner(双模型)、X-learner(交叉训练模型).结果显示,在体育消费随机对照实验中应用Uplift因果模型,可以基于基本模型进一步推导出各变量因素之间的因果关系,验证并分析自变量对因变量变化的影响;率先在体育消费市场研究与实验中应用Uplift因果模型可以填补我国体育消费实验数据分析方法的空缺.
AI Machine Learning:An Uplift Causal Model Study on Sports Consumption Experiment
The publication of Metallearners for Estimating Heterogeneous Treatment Effects Using Machine Learning in 2019 has sparked research enthusiasm for causal inference theory worldwide.Currently,many statistical models in machine learning and causal inference theory have been widely applied.This paper adopts research methods such as literature review,mathematical statistical analysis and case study to study the application scenarios of the well-known Uplift causal model in modern causal inference theory in sports.The Uplift causal models include S-learner(single model),T-learner(double model),and X-learner(cross training model).The results show that applying the Uplift causal model in a randomized controlled experiment of sports consumption can further derive the causal relationships between various variable factors based on the base model,and verify and analyze the impact of independent variables on the changes in the dependent variable;The application of the Uplift causal model in the research and experimentation of the sports consumption market can fill the gap in data analysis methods for sports consumption experiments in China.
AImachine learningmeta-analysissports consumptionUplift causal model
NETL
NSTL
万方数据
维普
