Hi , Thanks for the comment.  I had actually gone through the document , it very clearly explains the concepts of different storage however how do we specify the storage options is still missing. Is it a part of Cube designing process or is it interpreted by the OLAP server from the cube design ? Vishal ... View more

Hi, I have tried to work out using Sample data. data sample; input dates mmddyy8. a b; format dates mmddyy10.; datalines; 01/01/09    1   5 01/11/09 2   2 01/21/09 4 6 01/31/09 2 3 02/10/09 5 7 02/20/09 7 1 03/02/09 3 5 03/02/10 8 2 03/03/10 2 7 03/13/10 8 3 03/23/10 3 7 04/2/10    5 2 04/12/10 3 7 04/12/11 5 4 04/22/11 4 5 05/02/11 6 2 05/12/11 2 7 05/22/11 7 6 06/01/11 4 8 06/11/11 5 3 06/21/11 8 3 ; proc sort data = sample; by dates; run; /* Finding Unique dates more than 01/01/2010 which have atleast one year of history data*/ data unique_dates; set sample (keep = dates); by dates ; if dates  > = input("01/01/10" , mmddyy8.); if first.dates then output; run; /* Passing each date in the macro for finding correlation for that particular date based on data for last one year from that date*/ data _null_; set unique_dates; call symput("value",put(dates,mmddyy8.)); %macro_process(value); %put &value; run; %macro macro_process(value); data subset; set sample ; /* Subsetting Data for last one year from the particular date*/ if dates < = intnx('year',input("&value",mmddyy8. ),-1) then output; run; /*Running Correlation proc for particular date and relevant data*/ proc corr data = subset (drop = dates); run; %mend macro_subset(value); /*Each of the Correlation output can be saved in any form using ODS*/ Vishal ... View more

Hi, Create a macro which receives as input a date. This macro will be run once for each 1566 dates. In that macro first subset your data set with values relevant to the date. Like if date passed is 01 Jan 2008 , the subset should be all the records with date between 01 Jan 2008 & 01 Jan 2007. Then Use this temporary data set in your Correlation function. Then Use ODS for creating an excel file for the output of correlation. Include the macro in a loop which runs for each of the dates. Vishal ... View more

Ok, I think using a left join would solve the purpose select a.PATID , b.Therapy from PREG a left join THERAPY b on a.PATID = b.PATID I assume that Therapy data set has unique PATIDs. Please share the code you have written. Vishal ... View more

Hi, Tell me , if PAT id of two or more pregnancies (as in your example) are same , then both of them have same therapies , or there can be a case where PAT id is same for two pregnancies but therapies might be different. If therapies are same it can be done on SAS as well SQL. Vishal ... View more

Hi, Subquery is not required here , also it slows down the performance. Proc Sql; select distinct House_ID from have where House_ID <> '02' and Apple=1 ; quit; Just confirm the syntax for not equal to expression : <> Vishal ... View more

I have gone through this articles. It says we compare p-value with alpha , but the reason why alpha is chosen as the threshold value is not explained. ... View more