What's in Your Car?
The GKS Guide to Gas Mileage

�

With rising gas prices, we were interested in determining what variables tend to be associated with higher gas mileages.� In our data set, we had the following variables: Name of Vehicle, Sports, SUV, Wagon, Minivan, All Wheel Drive (AWD), Rear Wheel Drive (RWD), Suggested Retail Price (SRP), Dealer Cost, Engine Size, Number of Cylinders, Horsepower, City Miles per Gallon, Highway Miles per Gallon, Weight, Wheel Base, Length, and Width.�

Through previous analysis, we found several variables were highly correlated with one another.� SRP and Dealer Cost had a correlation of r = 0.9999.� Highway Miles per Gallon and City Miles per Gallon were also highly correlated (r = 0.941).� The number of Cylinders and Engine size have a correlation of r = 0.911.� Other fairly high correlations were also present.

Since the variables were highly correlated with one another, we used principal component analysis and factor analysis to reduce the dimensionality of our data set.� These results turned out to be quite interesting!� So sit back, relax, and enjoy the ride as you cruise through our super awesome findings...VRROOM!� VRROOM! � J

�Links

Principal Component Analysis


Factor Analysis

�2004 Car Data

The original data set can be found at http://www.amstat.org/publications/jse/datasets/04cars.dat
We deleted all vehicles with missing observations to do our analyses.