<u>Answer-</u>
D. No, the points are in a curved pattern.
<u>Solution-</u>
The difference between the observed or given value of the dependent variable and the predicted value is called the residual. Each data point has one residual.
<em>If the points in a residual plot are randomly dispersed around the horizontal or x-axis, a linear regression model is appropriate for the data. Otherwise, a non-linear model is more appropriate.</em>
Plotting a graph, by taking the residual values on ordinate and values of given x on abscissa, a non-random U pattern is obtained.
So a non-linear model will be a better fit for this data set.