[gov2001-l] seeing the log-likelihood iterations - Gov2001 - lists.gking.harvard.edu

List overview All Threads
Download

[gov2001-l] seeing the log-likelihood iterations

[gov2001-l] pset 6

[gov2001-l] obvious

hersh＠fas.harvard.edu

25 Mar 2009 25 Mar '09

5:44 p.m.

Hi: We are replicating an article that uses a random effects model. The authors show a likelihood ratio test that Stata reports in its output. A Google search taught me that the null model in this test "corresponds to the last iteration from Fitting constant-only model." Can I ask Zelig to show me such a thing? Thanks! Eitan

Reply

Show replies by date

plam＠fas.harvard.edu

25 Mar 25 Mar

5:49 p.m.

you can run a constant-only model in R by doing something like this for the formula: zelig(y~1, ...) 2009/3/25 Eitan Hersh <hersh at fas.harvard.edu>

Hi: We are replicating an article that uses a random effects model. The authors show a likelihood ratio test that Stata reports in its output. A Google search taught me that the null model in this test "corresponds to the last iteration from Fitting constant-only model." Can I ask Zelig to show me such a thing? Thanks! Eitan _______________________________________________ gov2001-l mailing list gov2001-l at lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

-- Patrick Lam Department of Government and Institute for Quantitative Social Science, Harvard University http://www.people.fas.harvard.edu/~plam

Reply

woolfalk＠fas.harvard.edu

6:02 p.m.

This just means that the LR test in Stata uses parameter estimates (and likelihood values) from two models -- the unrestricted model (e.g. the full model with all covariates) and the restricted model (which in this case seems to be a model where all parameters are set to zero and there is just an intercept term) -- to conduct the test. The "last iteration" probably just refers to the "last iteration" in the maximization algorithm -- so the "last iteration" gives us the parameter estimates that maximize the likelihood for each model. To do the LR test, all you really need from zelig is the likelihood value for the unrestricted and restricted models. Miya 2009/3/25 Eitan Hersh <hersh at fas.harvard.edu>

Hi: We are replicating an article that uses a random effects model. The authors show a likelihood ratio test that Stata reports in its output. A Google search taught me that the null model in this test "corresponds to the last iteration from Fitting constant-only model." Can I ask Zelig to show me such a thing? Thanks! Eitan _______________________________________________ gov2001-l mailing list gov2001-l at lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

-- Miya Woolfalk Ph.D. Student Harvard University Government and Social Policy

Reply

5523

days inactive

5523

days old

gov2001@lists.gking.harvard.edu

Manage subscription

2 comments

3 participants

tags (0)

participants (3)

hersh＠fas.harvard.edu
plam＠fas.harvard.edu
woolfalk＠fas.harvard.edu