Linear regression analysis (Robust regression, Stepwise Selection method)

• Consider the auto.csv below(mpg, displacement, horsepower, weight, acceleration, model year, cylinder, origin).

I want to build amultiple linear regression for predicting mpg as y variable. ( R or SAS, or any other software ok).

Could you please answer the question?

Thank you!

2)Consider all 5 variables (displacement, horsepower, weight, acceleration, and model year) except cylinder and origin, and the best main effects model obtained by stepwise selection method.

Use iteratively reweighted least squares (IWLS) approach to robust regression for dampening the influence of outlying cases.

Use Huber weight function and MAD (for calculation of scaled residuals).

The initial residuals may be obtained from the OLS model.

Carry out at least three iterations (excluding iteration 0) and compare the residuals, weights, and regression coefficients across iterations including the ones from OLS.

http://faculty.marshall.usc.edu/gareth-james/ISL/data.html