I discovered that three independent variables have standardized multiple coefficients (SMC) equal to 1.02 on a dependent variable. Are there any approaches to be considered for modifying a high R2 in a regression?

Similar questions and discussions