Howdy yall,   Does anybody know alternatives to the hazardratio statement that allow the statistician to specify different units for continuous variables?  This statement does not offer p-values.     ... View more

Is anybody here familiar with writing contrast statements for two-way interactions?  I have read the SAS Class, Contrast, and Hazardratio Statements document but im not sure how they came up with the design matrix especially the 1s and 0s.    Variables in the Model Design Matrix     β1 β2 β3 β4 β5 GROUP SEX groupA groupB sexMale groupAsexMale groupBsexMale A Male 1 0 1 1 0 A Female 1 0 0 0 0 B Male 0 1 1 0 1 B Female 0 1 0 0 0 C Male 0 0 1 0 0 C Female 0 0 0 0 0   What are the steps that I would follow to come up with a design matrix for a two-way interaction between two 2-level variables or any combination of variables for that matter? ... View more

Sas members,   does anyone know if it is possible to obtain p-values using the hazardratio statement?  Or are there alternative methods that allow the user to specify different units for continuous variables (that also come with the confidence intervals and p-values)?   ... View more

I am attempting to do a column same sorted by two variables with proc sql but get the error message below:   proc sql;  create table sales as select model, year sum(revenue) as sum_revenue, from carsales          _           22          76 Error 22-322: Syntax error, expecting one of the following: a quoted string, !, !!, &.... Error 76-322: Syntax error, statement will be ignored. group by model, year; run;   Have  model year revenue acura  2013  15000 acura 2014    30000 acura 2014    150000 hyundai 2012   300000 hyundai 2012 300000 bmw 2007 1000000 bmw 2007 1000000 bmw 2007 1000000   Want  model year revenue sum_revenue acura  2013  15000 15000 acura 2014    30000 180000 acura 2014    150000 180000 hyundai 2012   300000 600000 hyundai 2012 300000 600000 bmw 2007 1000000 3000000 bmw 2007 1000000 3000000 bmw 2007 1000000 3000000   Help?   ... View more