Analysis of Gas Mileage of a Car
Abstract
The objective of this work is to analyze a data set, Auto, from the R package ISLR: Introduction to Statistical Learning in R. The data set includes information for 392 observations on 9 variables including gas mileage, horsepower, weight in pounds, and engine displacement in cubic inches. The data set was taken from the StatLib library maintained at Carnegie Mellon University. The primary response variable will be gas mileage in miles per gallon, with all other variables serving as predictors, but other relationships with other response variables such as acceleration will be explored. Results were similar to expected; traits desirable for performance such as horsepower and engine displacement were negatively associated with gas mileage since performance usually comes at the cost of mileage. Additionally, weight of the vehicle was negatively associated with gas mileage and there was a positive association between year of manufacture date of the model and gas mileage. We produce a final multiple regression equation of mpg = -14.35 - 0.006632(wt) + 0.7573(yr). Additionally we produce two single variable regressions: mpg"=(-0.52+0.123(yr)^2 )^4 and ln(mpg) = 11.52- 1.058ln(wt).
Analysis of Gas Mileage of a Car
The objective of this work is to analyze a data set, Auto, from the R package ISLR: Introduction to Statistical Learning in R. The data set includes information for 392 observations on 9 variables including gas mileage, horsepower, weight in pounds, and engine displacement in cubic inches. The data set was taken from the StatLib library maintained at Carnegie Mellon University. The primary response variable will be gas mileage in miles per gallon, with all other variables serving as predictors, but other relationships with other response variables such as acceleration will be explored. Results were similar to expected; traits desirable for performance such as horsepower and engine displacement were negatively associated with gas mileage since performance usually comes at the cost of mileage. Additionally, weight of the vehicle was negatively associated with gas mileage and there was a positive association between year of manufacture date of the model and gas mileage. We produce a final multiple regression equation of mpg = -14.35 - 0.006632(wt) + 0.7573(yr). Additionally we produce two single variable regressions: mpg"=(-0.52+0.123(yr)^2 )^4 and ln(mpg) = 11.52- 1.058ln(wt).