Hi. If I do understand your need, you have to try the INTNX function (to shift 1 quarter back in the past) with the optional "END" fourth argument to move to the end of the period, that is the last day of the preceding quarter. [pre]last_quarter_month = INTNX("QUARTER", current_date, -1, "END") ;[/pre] This gives 31dec2010 for CURRENT_DATE to 01feb2011. If you only want the month, include the whole INTNX function into a MONTH function ; that will produce 12 as a result. Hope it helps. Olivier ... View more

Gregg is right : use slashes to go and read the next line. To avoid stopping on the first space, specify the informats this way (not just $ but $ followed by the maximum length you would expect). The : before the $ tells SAS to stop when it encounters the delimiter (specify one in your INFILE statement, because default is blank !) or the end of the line. [pre] input name :$100. / address :$20. / city :$50. st $2. +1 zip $5. / areacode $3. +1 phone $8. / star $10. focus :$50. ;[/pre] In this code, I use : when I don't know how long the values will be, but if I know the exact length (as for State, Zipcode of phone numbers) I don't mention any so that SAS reads exactly that number of characters. Hope it helps. Olivier ... View more

I don't think there is one, unfortunately. PROC FREQ is lacking the PRELOADFMT option. ... View more

Hi. In addition to Marius' answer, you have to use both NOTSORTED option in the FORMAT procedure and PRELOADFMT + ORDER=DATA in a TABULATE procedure to get exactly what you want. [pre]PROC FORMAT ; VALUE $ProvFmt (NOTSORTED) 'P2' ='BC' 'P3' ='AB' 'P5' ='SK' 'P1' ='ON' 'P4' ='MB' 'P11'='NT' 'P12'='YK' 'P13'='NU' 'P10'='PE' 'P8' ='NB' 'P6' ='QC' 'P7' ='NS' 'P9' ='NL' ; RUN ; DATA work.test ; DO i=1 TO 500 ; num = ROUND(RAND("UNIFORM")*12+1) ; product = CATS("P",num) ; OUTPUT ; END ; FORMAT product $provfmt. ; RUN ; PROC TABULATE ; CLASS product / PRELOADFMT ORDER=DATA ; TABLE product, N ; RUN ;[/pre] Regards. Olivier ... View more

Hi. Since SAS dates are stored as number of days (beginning on day 0 which is january 1st, 1960), what you want to do is basically a substraction. The SAS function called TODAY() or DATE() -- both give the same result -- will deliver the exact date every day. [pre]Data a; input date ddmmyy10.; gap = TODAY() - date ; cards; 05/02/2011 run;[/pre] ... View more

Hi Simone. Maybe a subquery would help : [pre]PROC SQL ; SELECT SUM(x) FROM ( SELECT DISTINCT id, x FROM myData ); QUIT ;[/pre] This is very similar to executing a PROC SORT with the NODUPRECS option prior to summing. Hope this helps. Olivier Message was edited by: Olivier ... View more

Peter, I do agree with you : having LAG work in any context would be really strange. That's why I was trying to use it in FCMP, expecting an error or warning message. But nothing came, except strange results. Moreover, when I run [pre] PROC FCMP LIST ; RUN ; QUIT ; [/pre] the Output shows : [pre] Listing of Variables used in the Program _N_ _ERROR_ _LIST_ _DELTA_ _DER_ _I_ The Program Lag Length is 0. [/pre] Does that mean that the "Lag length" can be tuned ? I didn't see any option about that. Maybe some SAS folk would have an answer. Thanks for your answers. Olivier ... View more

Hi. What you can do is change the template that displays the results of your Logistic procedure : try this code [pre] proc template; define table Stat.Logistic.ParameterEstimates; dynamic NRows; column Variable GenericClassValue Response DF Estimate StdErr WaldChiSq ProbChiSq StandardizedEst ExpEst Label; define Estimate; header = "Estimate"; parent = Stat.Logistic.vbest8; format = 20.8 ; end; end ; run ; [/pre] and then re-run your Logistic Proc. It will display values with 8 decimals in the Parameter Estimates table. Regards, Olivier ... View more

Hi. The reason why numbers are added at the end of the name is that SAS recalls that it has already given such a name. Its memory consists of catalogs in WORK (or wherever you said it was, using the option GOUT in your graphic procedure). If you run 9.1 or 8.2, the only requirement is to add at the beginning of your code [pre] PROC CATALOG CAT=work.gseg KILL ; RUN ; QUIT ; [/pre] It will clear all memory of creating images with the name you want to use. Testing the following code in SAS 9.2, [pre] ODS HTML GPATH="c:\temp" ; PROC GPLOT DATA=sashelp.class ; PLOT weight * height / NAME="widget" ; RUN ; QUIT ; ODS HTML CLOSE ; [/pre] it appeared that killing the GSEG catalog was not enough : there was a copy of the names in a HTML catalog, located in WORK either. That looks strange (because I don't remember SAS ever behaving this way... by the way, you 9.2 guys, am I missing something at this point ???), but anyway, when you get reed of BOTH catalogs, that works out fine. So try : [pre] PROC CATALOG CAT=work.gseg KILL ; RUN ; QUIT ; PROC CATALOG CAT=work.html KILL ; RUN ; QUIT ; [/pre] before your favorite G procedure to create images with the correct name. Regards, Olivier ... View more

Hi. The problem is, the way HTML handles image files is a problem here : when you see a picture in a Web page (any time), it is just a link to a separate image file. In Excel, the link from the HTML file to the image does not work very well to display (at least in my experience). In contrast, PDF includes both text & picture. That's why it works so fine in your example. What you can do is change the HTML destination for TAGSETS.EXCELXP, provided you run your programs with SAS version 9.1.3 at least, and read the files with Excel 2002 at least. It is designed to build REAL (almost) Excel files, which can include pictures. Regards. Olivier ... View more

Hi Jenn. You can use this shortcut : VAR var2: ; which stands for 'all variables having their name beginning with "VAR2" '. Regards Olivier ... View more

Hi. You can use the PERCENT informat in the INPUT statement. PS : the AVG function does not exist, its name is MEAN. [pre] data x; input id :PERCENT. sal :PERCENT. ; FORMAT _numeric_ PERCENT12.2 ; cards; 3.4% 7.8% 1.2% 8.9% ; run; data y; set x; y=MEAN(id,sal); run; proc print; run; [/pre] Olivier ... View more

Hi Vince. I think that you are lacking an ID statement in your Transpose procedure. This will allow SAS to connect newly created variables to Q2 values, instead of just re-arranging values by Q1 values as it currently does. Could you try this code and tell us if this is what you are looking for ? (I assumed that the output table you posted add the Q1 variable as the first column, and not "Q2", as Scott pointed out.) [pre] DATA test ; INPUT Q1 Q2 freq ; CARDS ; 1 1 2 1 2 2 1 3 3 2 1 2 2 3 1 3 1 2 3 2 2 3 3 3 ; RUN ; PROC TRANSPOSE DATA = test OUT = test2 ; VAR freq ; BY q1 ; ID q2 ; RUN ; [/pre] Regards Olivier ... View more

I think you would also achieve that using a simple Data step, with arrays, FILE and PUT statements, and VNAME (or VLABEL) function to retrieve the current variable name or label. In fact, this is how I create XML files, if it is for one-shot use. If not, I'd rather create a custom markup ODS destination using the Template procedure. Olivier ... View more