3
$\begingroup$

I plot a standardized residuals against the fitted values and it does exhibit a megaphone shape. It looks like there is more variation in the lower level of fitted values.

enter image description here

I then conduct a Breusch-Pagan Test for heteroskedasticity.

studentized Breusch-Pagan test

data:  reg4
BP = 87.88, df = 2, p-value < 2.2e-16

The p-value is quite small, which indicates that I should reject the null hypothesis and conclude heteroskedasticity. At this point, can I safely conclude I do face a heteroskedasticity and do some remedial measurements to deal with it?

Improve this question
$\endgroup$

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.