how to create a diagnostic plot in proc genmod dist=poisson link=log for model fit analysis?

[lj701712]

my model assesses the relative risk of positive_reaction (1/0) associated with therapy hours.

proc genmod data=mydata;

class id;

model positive_reaction = num_hours +other_vars /dist=poisson link=log;

output out=try predicted=p reschi=r resraw=resraw;

repeated subject=id /type=unstr;

quit;

proc sgplot data=try;

scatter x=p y=r;

scatter x=p y=resraw;

run;

i would like to create a diagnostic plot in the genmod procedure to see if this linearity relationship works, or i need to use cubic splines.

i  did the scatter plot of predicted and raw residual (resraw), it shows two straight lines

if i do the scatter plot of the predicted and pearson residual (reschi), it shows two distinct curved line

but i expect to see random dots which scattered the XY plain

i might have not done the correct residual plots.

what should i suppose to do?

thanks so much.

[StatDave]

See "Linearity in the logit (or link), testing" in the list of Frequently Asked-for Statistics (FASTats) in the Important Links section of the Statistical Procedures Community page. As suggested there, you can use the method shown in the "Nonparametric logistic regression" example in the GAMPL procedure documentation, or the ASSESS statement in GENMOD to check for linearity/adequacy of the specified form.

[Ksharp]

As StatDave said using the ASSESS statement in GENMOD to check right model.
https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.5/statug/statug_genmod_examples08.htm

https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.5/statug/statug_genmod_examples09.htm