Predict Miles Per Gallon (MPG) of Cars

Cheong Kok Hoe
25 Apr 2015

Background & Basic Data Analysis

The mtcars dataset in RStudio was extracted from the 1974 Motor Trend US magazine, and comprises fuel consumption and 10 aspects of automobile design and performance for 32 automobiles (1973-74 models).

The dataset contains

  • 32 observations.
  • 11 variables.

Building a Regression Model

The sample of the observations in the mtcars dataset are

          mpg cyl disp  hp drat   wt  qsec vs     am gear carb
Mazda RX4  21   6  160 110  3.9 2.62 16.46  0 Manual    4    4

Using am (Transmission) as predictor, wt(Weight) and hp(Horsepower) as confounders in the model.


Call:
lm(formula = mpg ~ am + wt + hp, data = mtcars)

Coefficients:
(Intercept)     amManual           wt           hp  
   34.00288      2.08371     -2.87858     -0.03748

Diagnostics of Regression Model

plot of chunk unnamed-chunk-4 Residuals vs Fitted Plot. Relationship is linear and variances of error are highly equal. Normal Q-Q Plot. Data points are normally distributed.

Using Model to Predict MPG of Cars

For example,

  • am = “Automatic”
  • wt = “3.5”
  • hp = “120”

Predicted MPG = 19.4

If you are ready, proceed to launch Predict MPG of Cars!