Hi, I am looking into some cost (Dependent Variable) for three (3) different procedure groups (categorical independent variable) and plan to present the mean of predicted cost for each of these three procedure groups. The independent variables will be procedure group, comorbidity, and variables for some socio demographic information. I used generalized linear model (proc genmod) with gamma distribution and log link. I learned SAS code for GLM model and wrote my SAS code as the following. I am not sure if I wrote SAS code in a right way to get result. Please correct me if anything is wrong. Thanks so much! (1) I put all categorical variables in the CLASS statement; (2) I put type3 in the MODEL statement to see if there is difference of cost among 3 procedure groups;      From results, it shows table of LR Statistics For Type 3 Analysis. I will go check the p-value for variable procedure_cat to see if there is significant different in the cost for these 3 procedure groups. Is it correct? (3) I used LSMEANS statement to have mean of predicted payment for each of procedure group, which is why I put variable procedure_cat after LSMEANS statement;      At the bottom of results, it shows procedure_cat Least Squares Means. I will present the mean for each of procedure groups in my table for the mean of predicted cost for each of procedure groups. Is it correct? (4) I put / ilink after LSMEANS statement as I saw someone suggested to do so since link=log was already defined, but I am still not clear about the reason of using / ilink; Proc genmod data=abc.mydata;                 Class  procedure_cat  female  race_cat  income_cat  comorbidity_cat;                 Model  payment = procedure_cat  age female race_cat  college  income_cat comorbidity_cat / dist=gamma link=log type3;                lsmeans procedure_cat / ilink; Run; ... View more

Hi, I am looking into some cost (Dependent Variable) for three (3) different kind of surgeries and plan to present the mean of predicted cost for each of three surgery groups. The independent variables will be surgery group, comorbidity, and variables for some socio demographic information. I think I should use generalized linear model (proc genmod) with gamma distribution. My SAS code is listed below. When I ran the code, there always were two same errors shown. I am not sure about the reason. I had nothing to do with SUBFUNC, and did not assign any library named SUBFUNC. Could you give me some hints? Another question is even though there were error shown up, I still got the result. Can I just use this result? Thanks so much! Proc genmod data=abc.mydata; ERROR: Libref SUBFUNC is not assigned. ERROR: Libref SUBFUNC is not assigned. WARNING: Unable to load prototypes from 'subfunc.cmplib'.                 Class  surgery_cat  female  race_cat  income_cat  urban  comorbidity_cat;                 Model  payment = surgery_cat  age female race_cat  college  income_cat urban comorbidity_cat / dist=gamma link=log ;                 Output out=abc.resultdata  prob=p; Run; proc means n mean stddev CLM data=abc.resultdata;      class surgery_cat;      var p; run; ... View more

Hi, I am working on some drug data set and need to specify patients using some drugs. The problem is we have some drug name but the data set contains not exactly same but similar name. The drug name in dataset may contain more word/letter than the name we have in hand. I just used exact match to identify patients using these drugs, so the result looks not correct. My code looks like the following. I just listed two names here, but actually I have many drug names. %macro drugs; %do i = 1 %to 10; data drug;      set drug; drug_1=0; drug_2=0; if brandname in ('ADOXA', 'VECTRIN')      or genname in ('DOXYCYCLINE HYCLATE', 'QUINACRINE')       then drug_1=1; if brandname in ('CIMZIA', 'SIMPONI ARIA')      or genname in ('ANAKINRA', 'GOLIMUMAB')        then drug_2=1; RUN; %end; %mend; %drugs; Too few patients were identified by the above code. I am thinking to use proc sql and like 'pattern' or =* operator, but not sure if it is a good way to do and also not sure if like or =* can be used with more than one drug names to do search. I am also concerning if sas can do macro and proc sql together. Could you help me write some sample code? Thanks so much! ... View more

Hi, I have a question about how to calculate day difference from the first date within a group (ID). Thanks much for your answer. My data looks like as below. There are two columns (ID, date) and some other columns, which I did not show here but I need keep. I need create a new variable date_diff to show day difference from the first date within each ID. I showed a example of ID=1 as below for how I need calculate. ID     date               date_diff 1     06JUL2006          0 1     23JUL2006     23JUL2006 - 06JUL2006 1     08AUG2007     08AUG2007 - 06JUL2006 2     03DEC2004 2     01JAN2005 ... View more

Hi, I have a question about update table using sql. My data sets are very large, so I would like to use proc sql instead of sort and merge. I have table a now and want to add in two more columns from table b. My first question is how to do update instead of create new table. My second question is how to deal with the character variable for linkage. Here is what I wrote for creating a new table. proc sql; create table new_a as      select a.*,                b. name_1, b.name_2      from a           left join              b      on a.num=b.num; quit; The variable [num] is character variable having many digit and first couple of digits may be zero. Log file noted data file b is in a format that is native to another host, or the file encoding does not match the session encoding. Cross Environment data access will be used, which might require additional CPU resources and might reduce performance. Thanks much! ... View more

Hi, I have a question about converting character variable to numeric variable in SAS. My data set contains a character variable, which is five (5) digits and each digit may be letter or number. Now I want to convert the variable to numeric (actually only the ones with all five digits work well to be convert to numerical, and also keep others with letter as they are used to be). How to combine this condition with input function or is there any other way to do? Thanks much! -Michelle ... View more