Dear Everyone,   I want to run the following codes for 31 files including S1 to S31 and want to get 21 output files including O1 to O31. I highlighted the locations of both input and output files in the code. I know I should use macro to perform the job. Could you please suggest me how should I write the macro code to solve the problem?   proc nlp noprint data=S1 absgconv=1e-15 OUTEST=T1; by obs; bounds R>=0, R<=1; min f; decvar R=1; f = abs(PRC1_IBES-(BVPS1_1+(((FEPS1_1/BVPS1_1_1)-R)/(1+R))*BVPS1_1 +(((FEPS2_1/BVPS2_1_1)-R)/((1+R)**2))*BVPS1_1_1 +(((FEPS3_1/BVPS3_1_1)-R)/((1+R)**3))*BVPS2_1_1 +(((FEPS4_1/BVPS4_1_1)-R)/((1+R)**4))*BVPS3_1_1 +(((FEPS5_1/BVPS5_1_1)-R)/((1+R)**5))*BVPS4_1_1 +(((FEPS6_1/BVPS6_1_1)-R)/((1+R)**6))*BVPS5_1_1 +(((FEPS7_1/BVPS7_1_1)-R)/((1+R)**7))*BVPS6_1_1 +(((FEPS8_1/BVPS8_1_1)-R)/((1+R)**8))*BVPS7_1_1 +(((FEPS9_1/BVPS9_1_1)-R)/((1+R)**9))*BVPS8_1_1 +(((FEPS10_1/BVPS10_1_1)-R)/((1+R)**10))*BVPS9_1_1 +(((FEPS11_1/BVPS11_1_1)-R)/((1+R)**11))*BVPS10_1_1 +(((FEPS12_1/BVPS12_1_1)-R)/(R*(1+R)**11))*BVPS11_1_1)); run; data T1; set T1; if _TYPE_='PARMS'; rename R=rgls1_1_1; keep obs R; run; data O1; merge S1 (in=xx) T1(in=yy); by obs; if xx and yy; run;   I appreciate your help. ... View more

Hello Everyone,   I used the following code more than 1000 times to merge and the code never lets me down. However, now I am having a weird error sign which I can no way solve.    My original code is as follows: proc sql; create table C1 as select a.*, b.* from connection a LEFT JOIN F1 b on a.ISIN = b.ISIN and a.year=b.fyear; quit;   I get the following log report:   Could you please help me in figuring out the problem? I appreciate your help.   ... View more

Dear Everyone,   I have the US patent data from 1926 to 2010. In each year, firms apply for patents in different technology class. I want to count the number of patents a firm applied for in each year from known and unknown technology class. Say, Firm 10001 (Permno- a unique firm identifier) applied for a patent 1000 (patnum- an unique patent identifier) in technology class 80 in year 1990 and then applied for another patent 1500 in the same technology class 80 in year 2000. Then I would assume that the firm filed for a patent in the known technology class in 2000. So the dummy variable identifying known technology class would take value 1 for patent 1500 of firm 10001 in year 2000.   Now assume that Firm 10001 applied for another patent 1600 in the technology class 90 in year 2000 for the first time in its history. So the dummy variable identifying known technology class would take value 0 for patent 1600 of firm 10001 in year 2000.   Could you please help in putting the above stated arguments in a proper SAS code? I would really appreciate your help. I have around 2 million observations in my original file from year 1926 to 2010. A firm may have multiple patent applications in each year and may not have a single application in a year.   All the variables- Year, Patnum, Permno, Class are numeric. Sample Data:   year Patnum Permno Class 1990 1000 10001 80 1995 900 10001 70 2000 1500 10001 80 2000 1600 10001 90 1926 1300 10002 70 2000 1400 10002 80 2010 1200 10002 90   Expected Outcome:   year Patnum Permno Class Known_Dummy 1990 1000 10001 80 0 1995 900 10001 70 0 2000 1500 10001 80 1 2000 1600 10001 90 0 1926 1300 10002 70 0 2000 1400 10002 80 0 2010 1200 10002 90 0 ... View more

Dear Everyone,    I am importing text file to SAS format. I want to use first row as name and start the data from third row. I am using the following code. It is not happening:   proc import datafile='H:\Seminar Class\Banking Seminar\Data by me\Bank Call Report\FFIEC data\FFIEC CDR Call Bulk All Schedules 03312011/FFIEC CDR Call Bulk POR 03312011.txt' out=POR0311 dbms=dlm replace; datarow=2; delimiter='09'x; run; ... View more