How can we use correlation analysis (Pearson text) to reduce the number of decision variables? (input variables).
There are cases where it may happen that the decision variables are correlated among themselves. Decision variables are correlated when they are a bit redundant. This correlation can be used to reduce the number of decision variables. The variables that are highly correlated can be removed from the model. For this we can use one if the criteria which is variance inflation factor(VIF).The decision variables having VIF more than 5 can be removed from the model.
Get Answers For Free
Most questions answered within 1 hours.