About tan1721

tan1721 · ‎02-17-2024

Hi, my post was marked as spam. could you please help @communities@sas.com. The title is "sample size for the categories of a nominal variable".

tan1721 · ‎02-17-2024

I have a nominal variable with 3 categories. I am running a GEE model as the response variable is repeated in 2 scenarios. Following is a table with the sample size for both scenarios. As shown in the Table, the sample size for category 3 is only 6 and 1 in scenario 1 and scenario 2, respectively. I searched in Google to find out whether it is possible to run multinomial GEE model with this small sample size but could not find anything. That is why I am reaching out to this community to learn if there is any requirement for sample size for every category of a nominal variable. Category 1 Category 2 Category 3 Total Scenario 1 18 10 6 34 Scenario 2 20 13 1 34

tan1721 · ‎09-27-2023

Thank you!

tan1721 · ‎09-27-2023

These are observation numbers. I have 34 observations from a driving simulator experiment. While did the analyses without these 2 outliers (2 and 26) and with these outliers, number of significant factors are higher when these 2 outliers are deleted. Considering this difference as substantial difference, should I include the results from both analyses?

tan1721 · ‎09-27-2023

I am fitting GEE model for a repeated measure dependent variable. Attached figure shows boxplots. My question is, is it necessary to delete the outliers (showing as star signs) to fit a GEE model with gamma distribution?

tan1721 · ‎09-24-2023

I am trying to fit GEE model and using "Proc GEE" method in SAS. This method gives "GEE parameter estimate" table. But "Proc GENMOD" also gives the "GEE parameter estimate". One difference I found is GENMOD procedure has the option to provide Wald statistics, but GEE procedure does not. Also, most of the journal papers in my filed (transportation) reported Wald statistics with their parameter estimate. Now, I am confused about two things: 1) should I use GENMOD procedure or GEE procedure? 2) should I report Wald statistics or Z statistic for GEE model?. I have been searching in internet, but was not able to find out the answer. Help from SAS community will be highly appreciated.

tan1721 · ‎08-30-2023

It worked. I had to increase the iteration numbers as well. Thanks!

tan1721 · ‎08-30-2023

Following is the histogram and Q-Q plot of deceleration data retrieved from driving simulator experiment. For this non- normally distributed data, I am using generalized linear mixed model to analyze the impact of gender and age on the deceleration rate. but the analysis is not working if I specify the distribution type as gamma, it is working only if I specify the data as normal. In that case, which distribution I should specify in GLMM? Does it become linear mixed model if the distribution is specified as normal in proc glimmix?

Online Status	Offline
Date Last Visited	‎02-26-2024 08:00 PM

Re: What to do when our spam filter traps your post

sample size for the categories of a nominal variable

Re: GEE procedure

Re: outliers in GEE

outliers in GEE

GEE procedure

Re: What distribution types should be specified for non-normal continu...

What distribution types should be specified for non-normal continuous ...

Re: GEE procedure

Re: GEE procedure

Re: What to do when our spam filter traps your post

sample size for the categories of a nominal variable

Re: GEE procedure

Re: outliers in GEE

outliers in GEE

GEE procedure

Re: What distribution types should be specified for non-normal continu...

What distribution types should be specified for non-normal continuous ...