[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Eva Poen <eva.poen@gmail.com> |

To |
Statalist <statalist@hsphsun2.harvard.edu> |

Subject |
Re: st: Opinions on fractional logit versus tobit - prediction and model fit |

Date |
Fri, 3 Apr 2009 22:00:58 +0100 |

<> Mike, many thanks for your comments, and the extensive list of references. Much appreciated. 2009/4/2 <carlto12@msu.edu>: > With a significant portion of the data piled up at these values it > may make sense to use a two limit tobit model. The jury is still out on this one, judging from the responses. Am I right in gathering that, although both models are probably misspecified, the glm version is more robust against misspecification and therefore worth pursuing? > While it is true that the > fractional models proposed by Papke and Wooldridge (1996, 2008) {2008 > Journal of econometrics: they extend the fractional response models to a > panel data setting} are quasi-likelihood does not mean that the predicted > values are not valuable. As long as you have correctly specified the > conditional mean function this implies that you have consistent, although > inefficient estimates. OK, thanks, I was starting to get confused with what I can validly interpet with this approach. > If your interest is in Average Partial Effects (APEs) > then the fractional response models will allow you to get consistent > estimates of these, regardless of the variance structure. While I am certainly interested in average partial effects, one of the main points of doing this is to learn more about the heterogeneity, and therefore the random effects themselves. I might end up doing some simulations, in order to asses the estimates that -gllamm- delivers. However, given that a single run on 2000 observations can take many hours to converge, this could be a very lengthy exercise... Thanks again, Eva > > Hope this helps, > Mike > > References: > > Gourieroux, C., A. Monfort and A. Trognon (1984), ?Pseudo Maximum Likelihood > Methods: Theory,? Econometrica, 52, pp. 681-700. > Papke, L. and J. M. Wooldridge (1996), ?Econometric Methods for Fractional > Response Variables with an Application to 401(k) Plan Participation Rates,? > Journal of Applied Econometrics, 11, pp. 619-632. > Papke, L. and J. M. Wooldridge (2008), ?Panel Data Methods for Fractional > Response Variables with an Application to Test Pass Rates,? Journal of > Econometrics, 145, pp. 121-133 > Chamberlain, G (1984), "Panel Data", Handbook of Econometrics. (Also on NBER > I think) > > > > Quoting "Eva Poen" <eva.poen@gmail.com>: > >> <> >> >> I'm looking at different ways to model my outcome variable, which is >> bounded between zero and one (zero and 20, actually, but I don't mind >> modelling the fraction). It's panel data, and I would like to model >> individual heterogeneity in the form of random effects (both random >> intercepts and random slopes). There are a lot of observations at zero >> and one, respectively. I'm reasonably confident that the random >> effects are independent of the other variables in the model. >> >> So far I have been looking at the fractional logit model, as >> introduced by Papke and Wooldrigde in their 1996 Journal of Applied >> Econometrics paper. I use -gllamm- to estimate a model with random >> effects. I have also been looking at the tobit model, which I again >> estimate using -gllamm- with random effects. >> >> I have a few doubts about the fractional logit model (FLM), and would >> like to hear other people's opinion: >> >> - Although it appears to be a very elegant solution, some people say >> that FLM is not well suited for problems with a lot of zeros or ones; >> for example, Maarten Buis said so in this post (but didn't provide a >> reference): http://www.stata.com/statalist/archive/2007-07/msg00786.html >> If someone knows any references where this is discussed, I'd be >> grateful to receive them. >> >> - Since FLM is quasi-likelihood, any likelihood-based approaches to >> model fit are ruled out. For the tobit model I can use those measures. >> The only other option I can think of for FLM is to compare predicted >> values with actual values. However, do predicted values in FLM make >> sense? We know that the distributional assumption is not true. So I'm >> wondering how meaningful predicted values are in this context. >> >> - I am getting sensible estimates for the random effects with the >> tobit approach, and not so sensible ones with FLM. In fact, FLM >> estimates two of the three to be zero. Is this a sign of my model >> being incorrectly specified, or could it be a sign of FLM not handling >> the zeros and ones very well? >> >> Many thanks, >> Eva * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

- Prev by Date:
**Re: st: Opinions on fractional logit versus tobit - prediction and model fit** - Next by Date:
**Fwd: st:reweight to compare two samples** - Previous by thread:
**st:reweight to compare two samples** - Next by thread:
**st: Update to -texdoc- available from SSC** - Index(es):

© Copyright 1996–2014 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |