Outliers and residuals Part 3 Q 11

Question 11 in Part 3 asks to detect outliers. We are trying to find out if any of the assumptions regarding Ordinary Least Squares regression have been violated. The easiest way to do this is to look at the standardised residuals. You can find the standardised residuals if you check the appropriate box when organising the regression. Let's look at this question:

 The dataset AUTO2  reports the price, horsepower, and ¼ -mile speed for 16 popular sports and GT cars.

a)   Find the estimated regression equation, which uses price and horsepower to predict ¼ -mile speed.

b)  Plot the standardized residuals against the predicted value y. Does the residual plot support the assumption about ?  Explain.

c)   Check for any outliers.  What are your conclusions? (15/42)

Below is the column of standardised residuals from the regression output for this question:

Look at the column, which is z-scores for each residual. We are looking for a z score which is larger than + 3 or smaller than - 3. There aren't any. So no outliers. I will not ask you to plot the standardised residuals against the predicted y value, which is what b) is asking.