Regression Homework

Regression models are very useful for creating predictive models. For this exercise, we will play with some different data yet again. I do this for two reasons. First, it gets a bit old to be using the same data set over and over again. Second, regression analysis relies upon the notion that you are predicting the functional relationship between variables and the various things measured in the classic Rice Center data set are not causal.

So instead, we will be using car data described as:

Cars were selected at random from among 1993 passenger car models that were listed in both the Consumer Reports issue and the PACE Buying Guide. Pickup trucks and Sport/Utility vehicles were eliminated due to incomplete information in the Consumer Reports source. Duplicate models (e.g., Dodge Shadow and Plymouth Sundance) were listed at most once.

And can be loaded into your session as:

library( MASS )
names( Cars93 )

 [1] "Manufacturer"       "Model"              "Type"              
 [4] "Min.Price"          "Price"              "Max.Price"         
 [7] "MPG.city"           "MPG.highway"        "AirBags"           
[10] "DriveTrain"         "Cylinders"          "EngineSize"        
[13] "Horsepower"         "RPM"                "Rev.per.mile"      
[16] "Man.trans.avail"    "Fuel.tank.capacity" "Passengers"        
[19] "Length"             "Wheelbase"          "Width"             
[22] "Turn.circle"        "Rear.seat.room"     "Luggage.room"      
[25] "Weight"             "Origin"             "Make"              

Manufacturer

Model

Type

Min.Price

Price

Max.Price

MPG.city

MPG.highway

AirBags

DriveTrain

Cylinders

EngineSize

Horsepower

RPM

Rev.per.mile

Man.trans.avail

Fuel.tank.capacity

Passengers

Length

Wheelbase

Width

Turn.circle

Rear.seat.room

Luggage.room

Weight

Origin

Make

The Question

The sole question for this topic is to have you figure out the best regression model that describes the variation in car weight in the Cars93 data set. In you evaluation of alternative models, make sure you are looking at the normality of predictors, the relative fit of the alternative models, etc.