Dear all, We use a mixed model approach to estimate the within and between variance. This works fine, unless the the within variance is larger than the between variance. We do not want negative variance components (i.e. the “nobound” option in SAS proc mixed). At the moment, we use JMP “variability charts” to obtain a Bayesian estimate of the within and between variance. What is the equivalent proc mixed code? Thanks a lot, Vaélrie ... View more

Dear all, Do you know whether it is possible to update conditionnally 2 fields in the same sql update statement ? This kind of synthax gives an error : proc sql &statusql noprint; update decava.person set death_date = (select iden_mortaldate from ident_sel_1 where ident_sel_1.iden_inss = person.social_security_nr) where exists ( select iden_mortaldate from ident_sel_1 where ident_sel_1.iden_inss = person.social_security_nr) , smals_last_update_date = (select iden_lastupdt_datetime from ident_sel_1 where ident_sel_1.iden_inss = person.social_security_nr) where exists ( select iden_lastupdt_datetime from ident_sel_1 where ident_sel_1.iden_inss = person.social_security_nr) ; quit; Without the where exists statements, an ORACLE error occurs. Any idea is welcome. Valérie ... View more

Dear graph experts, Is there a possibility in SAS to find connected components of a very large undirected graph and be able to draw this graph. Details on this theory can be found on Wikipedia at http://en.wikipedia.org/wiki/Connected_component_(graph_theory). Apparently there exist some tool to visualize those kind of graphs (GraphViz for example). Thanks a lot, Valérie ... View more