01-24-2018
ehdezsanabria
Obsidian | Level 7
Member since
06-21-2016
- 16 Posts
- 10 Likes Given
- 0 Solutions
- 1 Likes Received
-
Latest posts by ehdezsanabria
Subject Views Posted 4035 05-17-2017 03:50 AM 4062 05-13-2017 06:33 AM 4093 05-12-2017 06:01 AM 4132 05-11-2017 06:19 AM 1643 08-27-2016 01:38 PM 10132 08-09-2016 07:41 AM 2798 07-26-2016 11:05 AM 2798 07-26-2016 11:02 AM 2825 07-18-2016 02:07 PM 2834 07-18-2016 06:31 AM -
Activity Feed for ehdezsanabria
- Got a Like for Zero-inflated model using proc GLIMMIX. 03-04-2021 01:06 PM
- Posted Re: Non-est LSM in PROC MIXED for repeated measures (unbalanced data) on Statistical Procedures. 05-17-2017 03:50 AM
- Posted Re: Non-est LSM in PROC MIXED for repeated measures (unbalanced data) on Statistical Procedures. 05-13-2017 06:33 AM
- Liked Re: Non-est LSM in PROC MIXED for repeated measures (unbalanced data) for sld. 05-12-2017 06:03 AM
- Posted Re: Non-est LSM in PROC MIXED for repeated measures (unbalanced data) on Statistical Procedures. 05-12-2017 06:01 AM
- Posted Non-est LSM in PROC MIXED for repeated measures (unbalanced data) on Statistical Procedures. 05-11-2017 06:19 AM
- Tagged Non-est LSM in PROC MIXED for repeated measures (unbalanced data) on Statistical Procedures. 05-11-2017 06:19 AM
- Tagged Non-est LSM in PROC MIXED for repeated measures (unbalanced data) on Statistical Procedures. 05-11-2017 06:19 AM
- Tagged Non-est LSM in PROC MIXED for repeated measures (unbalanced data) on Statistical Procedures. 05-11-2017 06:19 AM
- Tagged Non-est LSM in PROC MIXED for repeated measures (unbalanced data) on Statistical Procedures. 05-11-2017 06:19 AM
- Liked Re: Non-normal data; PROC GLIMMIX for SteveDenham. 05-11-2017 03:48 AM
- Liked Re: Unable to calculate Type III fixed effects in proc GLIMMIX for SteveDenham. 10-26-2016 04:56 AM
- Posted Unable to calculate Type III fixed effects in proc GLIMMIX on Statistical Procedures. 08-27-2016 01:38 PM
- Posted Re: Zero-inflated model using proc GLIMMIX on Statistical Procedures. 08-09-2016 07:41 AM
- Posted Re: How to fit a multilevel model using proc mixed for longitudinal data on Statistical Procedures. 07-26-2016 11:05 AM
- Posted Re: How to fit a multilevel model using proc mixed for longitudinal data on Statistical Procedures. 07-26-2016 11:02 AM
- Liked Re: How to fit a multilevel model using proc mixed for longitudinal data for SteveDenham. 07-26-2016 11:01 AM
- Liked Re: How to fit a multilevel model using proc mixed for longitudinal data for Rick_SAS. 07-26-2016 11:01 AM
- Posted Re: How to fit a multilevel model using proc mixed for longitudinal data on Statistical Procedures. 07-18-2016 02:07 PM
- Liked Re: How to fit a multilevel model using proc mixed for longitudinal data for SteveDenham. 07-18-2016 02:03 PM
-
Posts I Liked
Subject Likes Author Latest Post 3 2 1 1 1 -
My Liked Posts
Subject Likes Posted 1 06-21-2016 08:24 AM
05-17-2017
03:50 AM
Hello again Sld, Thank you for your answer. I will continue reading, and I will ask questions once I finish understanding additional literature. I dislike wasting the time of others with nonsense questions. Best wishes, Emma
... View more
08-30-2016
02:37 PM
1 Like
Try the following:
PROC GLIMMIX DATA=new method=laplace empirical;
CLASS sample trt sym tpt;
MODEL Akkermansia= trt|sym|tpt/ link=log dist=negbin ddfm=bw OFFSET=loglib ;
random tpt/ subject=sample type=chol;
LSMEANS trt*sym*tpt/diff ilink;
NLOPTIONS maxiter=500 tech=nrridg;
RUN;
This fits a repeated measures design, where V1, V2 and V3 are repeated measures on the subject, which here is the sample. I think this more closely represents the data and the factors involved. This analysis points to a significant 3 way interaction, so you may want to slice your comparisons for each visit, to look at differences between treatments and whether these are mediated by symptom.
Steve Denham
... View more
08-09-2016
07:41 AM
Hello Steve, I am coming back to this answer because I still haven't got it right. I ran the model after doing the ln (x) transformation of the total number of counts (library). Now I understand what you meant with rescaling. I have realised that the values that I get are extremely small. My original data was given in percentages of the total number of counts. You suggested me to rescale it based on the units in which library size is expressed, but there are no units. It's just a number of counts. Is there any other way to express the means? Now that I see the results, the values are not representative any more (they are too small to be considered a percentage of the total). Again, thank you for your time! Emma
... View more
07-26-2016
11:05 AM
Dear Rick, Thank you for your advice, I will definitely try EFFECTPLOT to generate a figure like the one you showed: Thank you for the kind suggestion! Emma
... View more