在R中：如何在交互的预测变量上运行多元线性回归，而不对未交互的变量进行回归？

Question

For the life of me I cannot figure out how to run a multiple linear regression with two predictors interacted without also regressing on the interacted predictors by themselves. Here is an example:

When I use this script to do a regression

lagfit <- lm(formula = Production ~ DayOfWeek*Employees, data = train)

It returns a regression with three predictors: day of week, # of employees, and the interaction. But I only want to regress on the interaction. How can I do this?

Answer 1

您可以只使用:运算符代替* ：

lagfit <- lm(formula = Production ~ DayOfWeek:Employees, data = train)