Question

HI I trying to understand better the plots below.

Only have small sample size of 21 observation in data using. I not see anything drastic maybe but for thick tail in qqplot

Does this mean that linear model should not be use? Many thank.

I trying to better interpret these question, and really want to understand.

Please could you give feedback. Thankyou

Residuals vs Fitted Normal Q-Q 20◇ 2.5 3.0 3.5 4.0 4.5 -2 Fitted values Theoretical Quantiles Leverage plot of hat values Cook's distance 15 20 15 20 Obs. number

Answer 1

YES, you are correct in interpreting the dignostics plots for the regression model.

First plot - Residual vs Fitted

They are randomly distributed with reference to mean line of zero hence theyare not forming any constant pattern. Hence the Error Variance is remain constant.

Qqplot - The qqplot showing that the tail is thik but there may be chance that the regression fit is good.your data points are less hence the plot can be seen in that way.

But it is not the case that you can not use linear regression.

You may proceed with linear regression.