Comparing Poisson and Extra Poisson Models

Welcome to the forum for MLwiN users. Feel free to post your question about MLwiN software here. The Centre for Multilevel Modelling take no responsibility for the accuracy of these posts, we are unable to monitor them closely. Do go ahead and post your question and thank you in advance if you find the time to post any answers!

Remember to check out our extensive software FAQs which may answer your question: http://www.bristol.ac.uk/cmm/software/s ... port-faqs/
Post Reply
LiamMS90
Posts: 2
Joined: Sat Jul 28, 2018 1:56 pm

Comparing Poisson and Extra Poisson Models

Post by LiamMS90 » Sat Jul 28, 2018 2:30 pm

Dear Colleagues,

I am trying to compare a two-level Poisson model and a two level Extra Poisson model. From what I have read, in Mlwin this can only be done with a Wald test.

I was just wondering how to set up the Wald test or if anyone could point me in the direction of some instructional literature that might help? I have not been able to find any.

Many thanks,

Liam

ChrisCharlton
Posts: 1018
Joined: Mon Oct 19, 2009 10:34 am

Re: Comparing Poisson and Extra Poisson Models

Post by ChrisCharlton » Mon Jul 30, 2018 10:41 am

I asked Kelvyn Jones about this and he gave the following response:
If you want to do a Wald test in MLwiN – use the intervals and test window to test whether the additional term is different from zero.

Here is a simple NBD model - the extra term is associated with pi-squared
model.png
model.png (108.06 KiB) Viewed 38 times
And you want to test that the 0.839 is different from zero
That is you test bcons2.1
test.png
test.png (19.35 KiB) Viewed 38 times
And you see that p is very small so there is a difference – the test is quite approximate however and people often divide by 2 to get a one-tailed test.

LiamMS90
Posts: 2
Joined: Sat Jul 28, 2018 1:56 pm

Re: Comparing Poisson and Extra Poisson Models

Post by LiamMS90 » Tue Aug 07, 2018 11:37 am

Many thanks!

Liam

Post Reply