" information criteria can be used to select a best fitting covariance structure" https://communities.sas.com/t5/SAS-Statistical-Procedures/proc-glimmix-selection-stepwise-or-backwards/td-p/172976 automated is maybe not the best approach anywy, variable selection should be informed by clinical reasong and not entirely data-dependent, just my opinion
... View more
you would want to sort the dataset in descending order first "proc sort; by descending efficiency; run;" and then there would be a number of ways to select the first 15 obs of this dataset, eg just count obs and "proc print data=xxx (where=(count le 15)); var efficiency; run;"
... View more
a more powerful analysis could be a parametric, joint frailty model. Rogers and Pocock compare alternatives and declare the joint frailty to be a superior option, although they consider the case of recurrent events (which you don't have): https://www.ncbi.nlm.nih.gov/pubmed/24453096
... View more
not sure why you have two do loops and they use the same index "I". Why not just: %macro test(col); data ...; set .... (where=(colour=&col)) etc run; %mend; %test(WHITE); %test(BLACK); etc
... View more
datepart: http://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000245883.htm i don't know why you'd need to use the condition, extract the date, keep what you want, discard what you don't
... View more
you seem to have the right instinct eg you mention 'mixed modelling' and repeated measures anova but then you go astray. I did a quick google search, so this might not be the best example, but these data have the same structure as yours, thus their code is instructive: https://stats.idre.ucla.edu/sas/faq/how-can-i-perform-a-repeated-measures-anova-with-proc-mixed/ if you want to do something really 'simple' then maybe a friedman test
... View more
advice won't be definitive. personally i wouldn't be fixated on the 'repeated measures', you have 2 timepoints, the baseline is a baseline. If not for the missing data do proportional odds modelling (proc logistic, beware: https://digitalcommons.wayne.edu/cgi/viewcontent.cgi?article=1076&context=jmasm) at the primary timepoint. Could consider last-observation-carried-foward, although probably not because it couldn't be considered conservative in this scenario (assuming the primary timepoint is the last timepoint). Other options depend on the extent of drop out etc etc.
... View more
i would guess it's just an estimate statement with the exp option as follows estimate 'Exp(sex_cat)' sex_cat 1 / exp cl ; but make sure the results make sense
... View more
i don't understand your final sentence ie your Q. You just have to write code eg substr and index functions to extract the date data and then derive the new date variable
... View more
ok, i've been there before. I think i created tiff and then used gimp to convert to eps or pdf. Install gimp 2 then search google 'convert tiff to eps using gimp', something like that edit: http://gimpchat.com/viewtopic.php?f=8&t=8245
... View more
