Thanks ... View more

Yes, I found. Thank you so much ... View more

@baseballyanks1 : I think if you exchange the X and Y axes on this, it should be a little more apparent. Disease (=1) is associated with large values of logX, no disease(=0) with small values.  While there is some overlap, the disease group "stretches" out more in the positive logX direction from the group median than the no disease group.  Those points will drive the fit of the regression line, so that you get a fairly large slope.  By eyeball, I would estimate the slope at something like 9 (difference in logits / difference in logX) = 3.72/.4, which is really close to the ~10 value of the ML estimates.   SteveDenham ... View more

The error and solution are described here:   https://support.sas.com/kb/41/926.html ... View more

Thank you very much ^^  NZ got locked down again. I might try it next week. ... View more

Also see Maxim 28. Values stored in macro variables for later use in code rarely need a format. Formatting is only needed if the macro variable in question is used for human-readable display (e-g. in the text of a TITLE statement). In all other cases, the raw values are easier to use when referencing the macro variable. ... View more

The SAS documentation has an example of how to compute FIML by using structural equation modeling with PROC CALIS. ... View more

@sas_user4 wrote: The following code worked just fine:   data try; set new; if cmiss(of _all_) then delete; run;   Thanks! The above code will delete a row if any of the values are blank. I thought you wanted to delete a row of all of the values are blank. ... View more

If there are only two group, compare MEAN of two group by PROC TTEST or PROC N1PARM + willconx. compare PROPORTION(a.k.a risk ratio) of two group by PROC FREQ ; TABLE group*gender/relrisk ; run;   @Rick_SAS  wrote bolgs about these before. ... View more

You haven't shown the results of the test for the proportional odds assumption, which indicates whether you have a valid model. You might like to study the annotated output given here:   https://stats.idre.ucla.edu/sas/output/ordered-logistic-regression/ ... View more

Thanks for your suggestions. I've had a look at these, also trying to run Python using PROC FCMP, but this is all new to me.   https://blogs.sas.com/content/sgf/2019/06/04/using-python-functions-inside-sas-programs/   I also came across a group who are using this service, but setting it up looks pretty complicated!   https://ayup.agency/blog/2019/how-we-integrated-with-gov-uk-notify       ... View more

I think you have to specify a column:   https://documentation.sas.com/?docsetId=grstatgraph&docsetTarget=p03jtr6n56wu4ln1k77szqw1fjkc.htm&docsetVersion=9.4&locale=en   CURVELABEL="string" | column | expression specifies a label for the reference line or lines. Requirement If you use the COLN() or COLC() function in the X= or Y= option to specify multiple reference line intercepts, then you must use the COLC() function in the CURVELABEL= option to specify exactly one label for each reference line intercept. Otherwise, this option is ignored. See GTL Functions Used with the EVAL Function. Interactions If the X or Y argument specifies a value, then use "string". If the X or Y argument specifies a column, then use a column to define the label for each value. ... View more

Thank you! I noticed that paper. I can modify the bookmark using js, ghostscript or other tools after the PDF is created. However, I want to find if it can be down in SAS itself other than through third party solution because my PDF generated through SAS intrnet.  ... View more