Author(s)

Shi Haofeng, Zhang Jinyu, Wang Weiting

Corresponding Author

Shi Haofeng

ABSTRACT

As a superpower, the United States is in the leading position in the world in terms of military strength and social and economic development strength. American elections are held every four years, and 2020 is the year of the US presidential election. Republican candidate Donald Trump and Democratic rival Joe Biden are running for president. The purpose of this article is to study the impact of different candidates on the U.S. economy and the global economy through mathematical modeling. The focus of this study is to find and process data. In order to understand the objective impact of different candidates on the U.S. economy, after consulting a large number of data, we divided the US economic indicators into three levels: economic influencing factors, financial influencing factors and trade influencing factors, Under the three major influencing factors, six influencing branch indexes are determined. After classifying the data, we package different index data into the same table page, We put the data into Mpai (data processing platform), use XGBoost machine learning processing method to get the data we need, and clear the outliers, The method of outlier detection is 3Sigma outlier recognition. Finally, the data are fitted after regression analysis. After the fitting is successful, we can get the impact of different candidates on the U.S. economy according to the model analysis.

KEYWORDS

Economic impact, XGBoost machine learning, Data processing