Timeline for GLMM model specification help gender effects + an effect that is nested only within female
Current License: CC BY-SA 3.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 11, 2013 at 8:41 | comment | added | KalahariKev | @JakeWestfall Please see updated Question. Thanks. | |
| Nov 11, 2013 at 5:38 | comment | added | Jake Westfall | @KalahariKev What you posted in these comments is not enough for me to tell what you actually did. If you want further help, please edit your question to show me exactly what you did (all the code). | |
| Nov 11, 2013 at 3:41 | comment | added | KalahariKev | @JakeWestfall I have a bigger issue - I am doing something wrong with implementing lme(). Comparing models using your code "MvsF" with the values you specified, and using a factor variate "Gender" with only M and F inside, the model is exactly the same. Compare MvsF in previous comment to: GenderMale 13.01162 5.33739 2..438 0.014776, only the sign has changed. This has to be wrong, or why did I go through the dummy coding? This is the current lme call :dmd.nlme <- lme(dist~Age+MvsF+TempMax+MeanRain+herb1_dens+herb2_dens+YvsNY+herd3_dens, random = ~1|ANIMALID/Month, data = DMDdata) | |
| Nov 11, 2013 at 3:24 | comment | added | KalahariKev | @JakeWestfall this is output from summary.lme: MvsF -13.01162 5.33739 2.438 0.014776 * YvsNY 12.59318 4.54336 2.772 0.005575 ** | |
| Nov 11, 2013 at 3:23 | comment | added | KalahariKev | @JakeWestfall YvsNY 12.59318 4.54336 2.772 0.005575 ** | |
| Nov 11, 2013 at 3:23 | comment | added | KalahariKev | @probabilityislogic Would you care to expand on the dummy coding? Would you code it differently, or do you disagree with the interpretation of the final coefficients? Does your suggestion imply a single variable with three levels would be sufficient? I started to think that there is a way to recode Jake Westfall s variable MvsF to represent all three possibilities, but I personally would be unsure of the interpretation. | |
| Nov 10, 2013 at 8:00 | history | edited | Jake Westfall | CC BY-SA 3.0 |
typo
|
| Nov 10, 2013 at 7:26 | comment | added | probabilityislogic | Sounds like the OP should replace Gender and young present with female+young and female+no young. I disagree about your dummy coding comment though. | |
| Nov 10, 2013 at 3:23 | comment | added | Jake Westfall | @KalahariKev No problem. Let me know if you have any questions. | |
| Nov 10, 2013 at 3:13 | comment | added | KalahariKev | Hi Jake. Thanks for the answer. It has some intuitive sense, but I don't know enough to give it the thumbs up yet . I will try it while I wait to see if anyone upvotes your answer, or if another comes up. Cheers. | |
| Nov 10, 2013 at 2:51 | history | answered | Jake Westfall | CC BY-SA 3.0 |