Cook's Distance of Beta Regrssion

user_8503935 0
Viewed 98

I calculated Cook's distance manually and with the function cooks.distance with Beta Regression and I got two different results. Can someone please help me understand why?

Below is how I calculate Cook's distance (s):

require(betareg)
df<-data("ReadingSkills")
y<-ReadingSkills$accuracy
n<-length(y)

bfit<-betareg(accuracy ~ dyslexia + iq, data = ReadingSkills)
yhat<-fitted(bfit)
cook<-cooks.distance(bfit)
hatv<-hatvalues(bfit)
res<-residuals(bfit, type = "response")

RHS<-hatv/(1-hatv)

s23<-var(res)###
p23<-res^2/s23
Response<-(1/2)*p23*RHS

cbind(Response, cook)

As recamended here How to manually calculate Cook's distance I tried 3 inspite of 2 but same problem.

Thank you in advance for your help!

outliers regression r
1 Answers

Here main issuse is type of residual. I was using Response residual but betareg package used Pearson residual.

user_85039350
Related