Timeline for Likelihood and estimates for mixed effects Logistic regression
Current License: CC BY-SA 3.0
25 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 30, 2014 at 17:31 | vote | accept | Steve | ||
| Sep 2, 2014 at 11:43 | history | edited | Steve | CC BY-SA 3.0 |
|
| S Aug 27, 2014 at 11:31 | history | bounty ended | Steve | ||
| S Aug 27, 2014 at 11:31 | history | notice removed | Steve | ||
| Aug 26, 2014 at 9:44 | history | edited | Steve | CC BY-SA 3.0 |
fixed code; added results in the end
|
| Aug 25, 2014 at 20:39 | history | edited | Steve | CC BY-SA 3.0 |
updated progress with example code
|
| Aug 25, 2014 at 0:48 | comment | added | usεr11852 | Have you asked this in the R-sig-ME list ([email protected])? Some people might be able to help you there further. In addition: I would strongly urge you to study the paper Efficient Laplacian and Adaptive Gaussian Quadrature Algorithms for Multilevel Generalized Linear Mixed Models by Pinheiro and Chao. It contains results regarding AGQ performance as well as the linear algebra behind them. If you want to code them.... well, get ready for some serious sparse matrix decompositions. :D | |
| Aug 24, 2014 at 22:52 | comment | added | Glen_b | Steve, can you make sure that you agree with Randel's edit? You can re-edit or roll-back. | |
| Aug 24, 2014 at 22:45 | answer | added | Randel | timeline score: 3 | |
| S Aug 24, 2014 at 22:24 | history | suggested | Randel | CC BY-SA 3.0 |
made $L$ consistent and corrected the grammar
|
| Aug 24, 2014 at 22:13 | review | Suggested edits | |||
| S Aug 24, 2014 at 22:24 | |||||
| Aug 22, 2014 at 20:17 | history | edited | gung - Reinstate Monica | CC BY-SA 3.0 |
light formatting
|
| Aug 22, 2014 at 16:51 | comment | added | shadowtalker | You're right that it's not in the "classical statistics" toolkit, but that's for outdated computational reasons rather than conceptual, philosophical, or statistical reasons. In any case if you want to take a further look at it there are decent slides here: humis.bmi.utah.edu/humis/docs/organization_1863_1330728239.pdf and a mathoverflow question here: mathoverflow.net/q/41855 | |
| Aug 22, 2014 at 16:14 | comment | added | Steve | @ssdecontrol well, I'm not sure... But all the books I checked and the lme4,ordinal R packages, do not use MCMC. So, I would like to stick with that. At least at the beginning. | |
| Aug 22, 2014 at 16:00 | comment | added | shadowtalker | What's non-classical about MCMC to evaluate an integral? | |
| Aug 22, 2014 at 15:57 | comment | added | Steve | @ssdecontrol thanks, MCMC is good alternative. But I would like to apply the classical approach. | |
| Aug 22, 2014 at 15:41 | comment | added | shadowtalker | I don't have time to take a good look now but this looks like a good use for MCMC. | |
| Aug 22, 2014 at 14:25 | history | edited | Steve | CC BY-SA 3.0 |
Improved notation to be coherent (zeta became b and some other minor); updated with a step towards (hopefully) the solution
|
| Aug 20, 2014 at 13:32 | history | tweeted | twitter.com/#!/StackStats/status/502085878748299265 | ||
| S Aug 20, 2014 at 12:26 | history | bounty started | Steve | ||
| S Aug 20, 2014 at 12:26 | history | notice added | Steve | Draw attention | |
| Aug 20, 2014 at 9:32 | history | edited | Steve | CC BY-SA 3.0 |
updated question
|
| Aug 18, 2014 at 18:22 | history | edited | Steve | CC BY-SA 3.0 |
improved what I'm asking
|
| Aug 17, 2014 at 18:04 | history | edited | Steve | CC BY-SA 3.0 |
edited body
|
| Aug 17, 2014 at 13:28 | history | asked | Steve | CC BY-SA 3.0 |