tag:blogger.com,1999:blog-38545607.post2145066084388216109..comments2024-03-09T10:26:48.789+00:00Comments on economics, psychology, policy: Scobit: An Alternative Estimator to Logit and ProbitEmma Watsonhttp://www.blogger.com/profile/11466193733741012673noreply@blogger.comBlogger7125tag:blogger.com,1999:blog-38545607.post-4816435257064628522009-10-07T17:00:19.506+00:002009-10-07T17:00:19.506+00:00There's also:
Convenient Specification Tests ...There's also:<br /><br />Convenient Specification Tests for Logit and Probit Models, Davidson and MacKinnon<br /><br />http://ideas.repec.org/p/qed/wpaper/514.htmlMark McGovernhttps://www.blogger.com/profile/17133619200829364366noreply@blogger.comtag:blogger.com,1999:blog-38545607.post-21599154540148554702009-10-07T16:52:21.861+00:002009-10-07T16:52:21.861+00:00Thanks for flagging these papers Kevin. From what ...Thanks for flagging these papers Kevin. From what I can see, he looks at:<br /><br />(i) LM tests of mis-specification for ordered logit models <br /><br />(ii) score tests of normality in bivariate probit models <br /><br />http://ideas.repec.org/e/pmu76.html<br /><br />In relation to the Trinity of Testing Procedures (LR/LM/Wald), it seems that score tests are analagous to the LR approach, if I follow the Murphy paper correctly...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38545607.post-79791798398756975372009-10-07T09:26:31.983+00:002009-10-07T09:26:31.983+00:00There are a bunch of specification tests for probi...There are a bunch of specification tests for probit. Anthony Murphy published a few papers on this years ago, in <i>Economics Letters</i> I think.Kevin Dennyhttps://www.blogger.com/profile/17891633553910348880noreply@blogger.comtag:blogger.com,1999:blog-38545607.post-49273253043129551162009-10-06T13:19:10.357+00:002009-10-06T13:19:10.357+00:00...for scobes everywhere....for scobes everywhere.Anonymoushttps://www.blogger.com/profile/06818173517378776252noreply@blogger.comtag:blogger.com,1999:blog-38545607.post-67919925174091838872009-10-06T11:50:24.764+00:002009-10-06T11:50:24.764+00:00Invented by Scoby?Invented by Scoby?Mark McGovernhttps://www.blogger.com/profile/17133619200829364366noreply@blogger.comtag:blogger.com,1999:blog-38545607.post-68584180849352793612009-10-06T10:16:59.408+00:002009-10-06T10:16:59.408+00:00I've used the scobit (or skewed logit) to test...I've used the scobit (or skewed logit) to test for the presence of skewness in a logit model (using a likelihood ratio approach).<br /><br />I'm not aware of how to test for non-normality in a probit or logit model (or if it is in fact possible). However, (ignoring kurtosis) using the scobit to test for skewness may be the best that we can do.<br /><br />I use heteroskedastic probit in a similar fashion, to test for the presence of heteroskedasticity in a probit model.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38545607.post-11813777494939144782009-10-05T19:01:14.182+00:002009-10-05T19:01:14.182+00:00I think I tried scobit once (having stumbled on it...I think I tried scobit once (having stumbled on it in Stata). It didn't make much difference but its nice to have alternatives to the usual suspects. In general I prefer probit because it leads more easily to generalizations.Kevin Dennyhttps://www.blogger.com/profile/17891633553910348880noreply@blogger.com