Eason Miao
Master Student from Rice University
This Slide details a simple linear prediction app at this link. This simple app tried to predict the average mpg value given car weight and transmission type. You can open this link and tried to input weight value and select transmission type to see the predicted result.
From the below figures, we can easily detect a difference of mpg between different transmission type. And those manual type have larger mpg value than automatic. Also we can see clearly a linear relationship between car weight and mpg. Since those two variables are both related to mpg, we include them all in our model.
Based on the summary, we can see that transmission type, weight and their iteraction all have very high siginificant values. And latter when we visualized the fitted lines, we should pay attention their different slopes and intercepts. Also from the summary, we can tell that manual transmission under normal condition is much fuel saving than automatic.
fit1 <- lm(mpg ~ am * wt, data = mtcars)
summary(fit1)$coef
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 31.416 3.0201 10.402 4.001e-11
## am1 14.878 4.2640 3.489 1.621e-03
## wt -3.786 0.7856 -4.819 4.551e-05
## am1:wt -5.298 1.4447 -3.667 1.017e-03
We can see that the two types of points fit well. And from the residuals plot, we observe most of the residuals centered around the 0 horizontal. Although this is not a perfect model, please visit link and try its prediction result! Please feel free add my linkedIn. Thank you!!
