When it comes to Evaluation of Linear Regression which Evaluation Metrics would be the best when we have redundant Variables in our dataset?
Question
Suppose we are having a number of Redundant Variables in our dataset which are not of any use and we are considering them in our model prediction process For Linear Regression so which metric(R-Square or Adjusted R-Square) will give us the proper evaluation of accuracy??
in progress
0
Interview Question
4 years
1 Answer
942 views
Newbie 0
Answer ( 1 )
Adjusted R-Squared will be more Suitable for the evaluation because in R Square though we have Redundant Features the value of R-Square will not decrease with increase in Features which are not useful while prediction. It remains Same or increases it value. Adjusted R Square value increases only when the variable in the model adds value to it. So, it is better to use Adjusted R-squared when there are multiple variables in the regression model. This would allow us to compare models with differing numbers of independent variables.