Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

LL of the Null model - CBC/HB

Hi,

I'm trying to determine the LL of the null model for a HB run model. I understand from other posts that the model LL is determined by:

A) For each respondent, take the natural log of the RLH and multiply it by the number of tasks completed by that respondent: LN(RLH)*#Tasks.  This gives you the LL per respondent.

B)  Multiply that amount by the number of respondents, giving you the total LL across respondents.

However, I don't know how to determine the LL for the null model. I want to find this out to determine a McFadden's R^2.

Keith mentions in another post that Perct.cert is the same as McFadden's R^2 but with values in the. 5-.7 or higher range this would make the model fit extremely good as McFadden's R^2 in the range of .2-.4 represent values of extremely good fit. So I'm a little confused.

I'd love some direction here.

Jasha
asked Nov 7, 2017 by Jasha Bowe Bronze (1,745 points)

1 Answer

0 votes
Jasha,

The model fit of .2 to .4 is from an old paper about aggregate logit models, not respondent-level models - I don't usually see people compute McFadden's rho-squared for HB models, so I'm not sure what my target would be there.

The LL for the null model is easy.  If each of your questions has, say 4 alternatives, then a null model with all utilities=0 would choose by chance alone and you'd expect to get 25% correct.  

Now follow steps A and B in your post, except use 0.25 instead of RLH and you'll have the null model's log likelihood.

If you have varying numbers of alternatives across sets or a dual response none that contributes to your model then the null LL calculation will be more complicated.
answered Nov 7, 2017 by Keith Chrzan Platinum Sawtooth Software, Inc. (62,500 points)
...