[vsni.co.uk] Contact us | Site map
Forum | VSN International Forum Index
Author Message
Post new topic

<  GenStat  ~  HGLM

WRIGHT Carole
Posted: Thu Jan 05, 2017 12:42 am Reply with quote
Guest
Happy New Year

I apologise in advance that the questions in this email are more theoretical than GenStat based. I have looked through old GenStat course notes and Lee, Nelder and Pawitan (2006), but would appreciate some guidance.

I am wanting to do a probitanalysis of mortality data where I have a random replicate term. Previously there has been no need for a random term so I use the PROBITANALYSIS procedure. From the analysis I want to be able to obtain estimates of the 99% lethal dose (LD99). Using a binomial GLMM I don’t seem to be able to get the estimates using the FIELLER procedure. I found an updated MFIELLER procedure in the mailing list archives but still had no luck.

I then tried a HGLM and managed to obtain what appears to be sensible output and also the LD99 with fiducial limits (using the updated MFIELLER). Yah!

However, I’m not confident with HGLMs and therefore have a few questions for you.

<![if !supportLists]>1. <![endif]>Previously using PROBITANALYSIS we fit logit, probit and complementary log-log links and use various criteria to determine the ‘best’ model. From the HGLM output is it the -2 ´ Pv(h) likelihood statistic that I can use to compare the different link functions for the fixed model?

<![if !supportLists]>2. <![endif]>I confess to having done an HGLM course several years ago with Roger and in my notes, with regard to the dispersion parameter, I have scribbled “trying to explain dispersion by the random terms so keep it fixed”. I’m not sure if this was relevant just to the example in the notes or was a general statement. In what instance would you let the dispersion parameter be estimated?

As I am sure you are all still reminiscing about the recent festive season, I will limit my questions to these two primary areas. I appreciate any responses received and wish you all well in 2017.

Kind regards
Carole

------------------------------
The information in this email together with any attachments is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. There is no waiver of any confidentiality/privilege by your inadvertent receipt of this material.
Any form of review, disclosure, modification, distribution and/or publication of this email message is prohibited, unless as a necessary part of Departmental business.
If you have received this message in error, you are asked to inform the sender as quickly as possible and delete this message and any copies of this message from your computer and/or your computer system network.
------------------------------

Post generated using Mail2Forum (http://www.mail2forum.com)

Display posts from previous:  

All times are GMT
Page 1 of 1
Post new topic

Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You can attach files in this forum
You can download files in this forum