THere is a way of coding PROC LOGISTIC that goes
MODEL Event/Trials=...
This uses special variables refering to how many total cases there are (I am not sure how to create this variable in SAS). Effectively it aggregates the data. Unfortunately I am not familiar with this syntax and it is used in SAS for some key analysis notably overdispersion where corrections exclusively use this command style.
Does anyone know how to do the following without the event/trial format?
MODEL Event/TRIAL=IV1 IV2/Scale=p (or alternately SCALE=WILLIAMS)
I am not sure these corrections for overdispersion are possible without this format
MODEL Event/Trials=...
This uses special variables refering to how many total cases there are (I am not sure how to create this variable in SAS). Effectively it aggregates the data. Unfortunately I am not familiar with this syntax and it is used in SAS for some key analysis notably overdispersion where corrections exclusively use this command style.
Does anyone know how to do the following without the event/trial format?
MODEL Event/TRIAL=IV1 IV2/Scale=p (or alternately SCALE=WILLIAMS)
I am not sure these corrections for overdispersion are possible without this format