www.cmm.bristol.ac.uk/forum

Hi,

I am running a 2-level random slope cumulative logit model using R2MLwiN (level 1: individual, level 2: country). I understand that when interpreting the estimated coefficients of the model, we have to reverse the sign of coefficient before transforming it back to odds ratio.

As I need to examine the u0 (random intercept) and u1 (random slope) of each country, I am wondering if it is also necessarily to reverse the sign of u0 and u1? I want to plot two caterpillar plots for u0 and u1 as well as the following predicted function: xb + u0[model@data$country] + u1[model@data$country] * model@data$var1.

Thank you very much.

Regards,
Vivian

Hi Vivian,
Thanks for your question. I wonder whether you can clarify a bit? When you talk about reversing the sign of the coefficient is this because of the base category you have chosen in the cumulative logit? I.e. if you are using the lowest category as base then the values for the thresholds would be -1* the values with the highest category for base.
If this is the case then to translate any other parameters i.e. fixed effects and residuals to the alternative ordering you will indeed swap the signs.
Hope this helps,
Bill.

Thansk Billb. Your solution helps me.

Thanks Billb, your answers clarify my questions!

No worries both. Basically if the probability of being in categories k and upwards is p then the probability of being in categories k-1 and downwards will be 1-p and if logit(p) = a then logit(1-p) = -a hence when you switch the base category then the main change will be a switching of signs for all the coefficients.
Best wishes,
Bill.