About Saraja

Saraja · ‎02-20-2019

Let me ask this question on the author's behalf so that we will have some meaningful answered down the line: We have multiple decors by the ID. There are missing values due the fact that there are multiple records per ID instead on unique values. ID age2015 age2016 BP2015 BP2016 Weight2015 Weight2016 1 55 235 178 1 56 201 154 Using ID as a grouping variable how we can merge both records into single one so that there will be no more missing values such as here: ID age2015 age2016 BP2015 BP2016 Weight2015 Weight2016 1 55 56 235 201 178 154 Also, there are several hundred such variables, can we run the code without naming the variables explicitly. Thanks! Sartaj

Saraja · ‎06-18-2018

I am running PROC GLM for the fixed effects regression of variable &depVar. over independent variables &indVar., while controlling for fixed effects for the geographic variation HRR_x_Year. Both ways, the results must be equivalent; however, in first case it complains about a singularity and R-square jumps to 0.84, where as in later case the R-square due to HRR_x_Year only changes about 0.06.Is the implementation different intrinsically. PROC GLM DATA=WORK.MODEL2; absorb HRR_x_Year; model &depVar.= &indVar. / SS3 SOLUTION ; run; "Note: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whose estimates are followed by the letter 'B' are not uniquely estimable." PROC GLM DATA=WORK.MODEL2; Class HRR_x_Year; model &depVar.= &indVar. HRR_x_Year / SS3 SOLUTION ; run;

Saraja · ‎03-10-2018

Thank you very much Art, this code is very elegant. I was hoping to put the wild card terms into the one list as in the last line but looks like that its not possible. Again, thanks !

Saraja · ‎03-10-2018

Hi, I am new to Proc Sql and wondering how to use wild cards such as 'SAS%' (SAS*) or 'SAS____ to identify the library names with SAS as prefix in the following code. Same is treu for the MAPS libraries as prefix or SAS as suffix. Can someone explain these wild cards by showing side by side how they are different from data step. Appreciated! data metadata2; set sashelp.vcolumn; where libname not in ('SASHELP', "SASHELP", 'SASUSER', 'SAMPSIO', 'SUBFUNC', 'MAPS', 'MAPSSAS', 'MAPSGFK', 'SAL428SL', 'STP:', 'TLV4'); run;

Saraja · ‎10-04-2017

I understand that based on the surveyselect proceedure, the output file DATA_SAMPLE has the 1000 replicate samples of size 0.025 of given data DATA_FULL which is still 2.5 times bigger then the original data file. I don't understand how SAS will know which replicate sample to chose without parsing the data, or without using the Replicate variable with BY statement, or without a loop. I am running similar models and the problem persists after the fix; I will appreciate if you explain this a bit more. Secondly will you kindly also explain what you mean by ODS OUTPUT TABLE1=DS1 TABLE2=DS2 ETC and how we can actually impliment if we don't know the exact number of tables we will get.

Online Status	Offline
Date Last Visited	‎01-04-2023 02:11 AM

Re: sas sql : group by and missing values

PROC GLM, obsorb vs. class statement: Results are different

Re: Wild Cards used in Proc SQl are different that those used in Data ...

Wild Cards used in Proc SQl are different that those used in Data Step...

Re: GLIMMIX for repeated measures and complicated modeling problem

Title1; number of lines

Re: CASE WHEN Statement

Re: Wild Cards used in Proc SQl are different that those used in Data ...

Re: PROC GLM, obsorb vs. class statement: Results are different

Re: proc glm: class vs. absorb

Re: sas sql : group by and missing values

PROC GLM, obsorb vs. class statement: Results are different

Re: Wild Cards used in Proc SQl are different that those used in Data ...

Wild Cards used in Proc SQl are different that those used in Data Step...

Re: GLIMMIX for repeated measures and complicated modeling problem