www.cmm.bristol.ac.uk/forum

Dear Mahadev11,

Suggest you calculate VPCs based on the latent response formulation of the logistic regression model. See Module 7 of the LEMMA online course (Section 7.2) or any decent multilevel textbook for details.

Best wishes

George

Dear rahulvbb,

The predicted state random effects measure how much higher the predicted log-odds are in each state relative to the average state.

Researchers often plot these in a caterpillar plot to aide graphical interpretation. You can do this using the -serrbar- command.

Best wishes

George

Hi Nils, Your DGP and runmlwin syntax all look fine. The problem appears to be with some particularly high counts generated by your DGP. Essentially if you specify a lower true value for your level-2 variance everything is fine, but as you pump up the level-2 variance you get more and more extreme c...

Dear Jack, Given a binary response, the model looks rather complex with many random coefficients. A binary response contains far less information than a continuous response. Also your cluster sizes are small with 8 pupils per school on average. There will be little variation in the response within c...

Dear Francisca, An estimate of zero for the level-4 variance looks a bit strange. You don't appear to have done anything obviously wrong in terms of your runmlwin syntax so the answer probably lies with your data. It would be worth trying to fit the model by MQL1, MQL2, PQL1 and especially MCMC to c...

Dear francescomolteni , I'm afraid there is no automated way to do this. You will have to do the calculations by hand from first principles. Module 7 of the lemma training materials and in particular the accompanying runmlwin do-files may be of some help. See for example section 7.5 http://www.brist...

Hi Mahadev11, The easiest approach would be to use the latent response method. Any good multilevel text book will describe this. You could also look at our free online course Module 7: Multilevel Models for Binary Responses STATA Practical (p.18) Essentially you calculate the VPC statistics in the u...

Dear csswader2 , Suppose you want to perform a Wald test of the current model against a simpler version where you omit the last two covariates in your model, fatalism and informult. This is equivalent to testing the null hypothesis that both coefficients are jointly equal to zero H0: \beta_9 = 0 &am...

Hi, The models look fine from a purely runmlwin perspective, but I'm afraid I can't comment as to whether they are equivalent to the split population models to which you refer. If you can write your desired model down as a model equation we can then have a think as to whether it can be implemented i...

Hi davydzombre, These days people almost exclusively use the latent response approach to calculating ICC/VPC coefficients. However, there isn't a latent response formulation of the Poisson model so you can't do this. What you propose is a simple statistic to summarize the proportion of higher-level ...