Biplots of Multivariate Data Guided by Linear and/or Logistic Regression Huh, Myung-Hoe; Lee, Yonggoo;
Linear regression is the most basic statistical model for exploring the relationship between a numerical response variable and several explanatory variables. Logistic regression secures the role of linear regression for the dichotomous response variable. In this paper, we propose a biplot-type display of the multivariate data guided by the linear regression and/or the logistic regression. The figures show the directional flow of the response variable as well as the interrelationship of explanatory variables.
Data visualization;biplot graph;linear regression;logistic regression;dimensional reduction;