What is stepwise regression,and when is it desirable to make use of this multiple regression technique?

Stepwise regression is a multiple regression estimation technique whereby independent variables are added and deleted one at a time to/from the regression equation.The decision to add or delete a variable is made on the basis of whether that variable improves the model.The use of stepwise regression can reduce the possibility for multicollinearity since it is unlikely that two highly correlated x variables will be included in a multiple regression that is estimated using the stepwise technique.This technique is useful when there are a great many independent variables.