About Christobal

Christobal · ‎04-21-2016

Oh yes. This is really the moral that I have to accept 😉 For the future I will pay more attention on this. 🙂

Christobal · ‎04-21-2016

Good morning everybody. First of all thanks for your replies. They were very helpful. And with the hint of FreelanceReinhard I was able to solve the problem. With the -lrecl max option I could solve it. THANK YOU VERY MUCH FOR THIS. I did not know this before. My fault was that I did not really follow the warning "WARNING: The quoted string currently being processed has become more than 262 characters long. You might have unbalanced quotation marks." because I read this article http://support.sas.com/resources/papers/proceedings11/262-2011.pdf and there I understood that this message can be more or less ignored. So thanks again to you for the help. Best regards Christobal

Christobal · ‎04-20-2016

Hello all, after a while I am writing again because I have a very strange problem: I have a program that is started from a scheduler via a batch file: The batch file contains the following line: start "Titel" "C:\Program Files\SAS93\SASFoundation\9.3\sas.exe" "C:\Users\adm\Desktop\LParcheck.sas" This solution was running for many many years, but after a few weeks the start of the sas program via batch file does not work any more. BUT when I open my SAS Base and run the program from there everything is running fine. Only when I start the sas program via the above mention bat file I receive an error. This is the error that is written in the log after starting it via bat file (please see excerpt from log file): 388 quit; 389 390 %macro mail; 391 proc freq data=work.lparliste noprint; 392 tables newdate /out=work.dat; 393 run; 394 395 data work.anzahl; 396 set work.dat; 397 keep Datum Anzahl; 398 Datum=newdate; 399 Anzahl=count; 400 run; 401 402 ods html close; 403 filename temp email to="X@tde" 404 CC="CC@Xde" 405 from="X@.de" 406 subject="Qualitätsmanagement Unix-Lpars Std-Ebene" 407 type="text/html"; 408 409 ods html body=temp rs=none; 410 411 proc report nowindows data=work.delta headline; 412 Title 'Lpars, zu denen keine Daten gesammelt werden können:'; 413 label lpar= 'Lpar-Name:'; 414 run; 415 416 417 ods html close; 418 419 %mend mail; WARNING: The quoted string currently being processed has become more than 262 characters long. You might have unbalanced quotation marks. 420 421 %macro checkit; 422 /* Anzahl member bestimmen;*/ 423 %LET anzmem = 0; 424 %PUT &anzmem; 425 DATA _NULL_; 426 SET POINT=X work.delta nobs=anzmem; 427 CALL SYMPUT('anzmem',anzmem); 428 STOP; 429 RUN; 430 431 %put &anzmem; 432 %if &anzmem > 0 %then %mail; 433 run; 434 %mend checkit; WARNING: The quoted string currently being processed has become more than 262 characters long. You might have unbalanced quotation marks. 435 436 %checkit; 22 Das SAS System 14:00 Wednesday, April 20, 2016 0 0 _________________________________________________________________________________ 22 ERROR 22-322: Syntax error, expecting one of the following: ein Name, eine Zeichenkette in Hochkommata, eine numerische Konstante, eine Datetime-Konstante, ein fehlender Wert, (, +, -, BTRIM, CALCULATED, CASE, INPUT, PUT, SUBSTRING, TRANSLATE, USER. _________________________________________________________________________________ 200 ERROR 200-322: The symbol is not recognized and will be ignored. NOTE: PROC SQL set option NOEXEC and will continue to check the syntax of statements. NOTE: The SAS System stopped processing this step because of errors. NOTE: PROZEDUR SQL used (Total process time): real time 0.11 seconds cpu time 0.01 seconds ERROR: Errors printed on page 22. NOTE: SAS Institute Inc., SAS Campus Drive, Cary, NC USA 27513-2414 NOTE: Das SAS System used: real time 0.78 seconds cpu time 0.38 seconds Does anybody have an idea what could be the reason for this problem? I am totally helpless after searching Google for many hours. Thanks in advance and best regards Christobal

Christobal · ‎03-31-2014

Hello Chris, thanks for your answer. The "overlay" option was the keyword. Now it works 🙂 Hello Udo, I will also try your option with proc sgplot, as it looks very interesting and seems to help for the future 🙂 Best regards

Christobal · ‎03-25-2014

Ok, now I calculated my trend and my table looks like this: Month Units Prediction pctChange trendPct Trend 2013/06 200000 0,38% 2013/07 210000 204000 5,00% 0,38% 2013/08 198000 208000 -5,71% 0,38% 2013/09 202000 212000 2,02% 0,38% 2013/10 210000 216000 3,96% 0,38% 2013/11 179000 220000 -14,76% 0,38% 2013/12 195000 224000 8,94% 0,38% 2014/01 198000 228000 1,54% 0,38% 2014/02 202000 232000 2,02% 0,38% 2014/03 236000 202767,6 2014/04 240000 203538,117 2014/05 244000 204311,562 2014/06 248000 205087,946 2014/07 252000 205867,28 2014/08 256000 206649,576 2014/09 260000 207434,844 2014/10 264000 208223,096 2014/11 268000 209014,344 2014/12 272000 209808,599 Now I want to print this table in a pdf document. Till today I only printed the "units per month" and the "prediction per month": proc gplot data=work.unitsvsprediction; ; plot units*monat=4 / vaxis=axis1 haxis=axis3 grid; plot2 prediction*monat=2 / vaxis=axis2; run; I tried to extend this statement by plot3 trend*month.... but unfortunately proc gplot doesn't support 3 plots. Is there any other possibility to solve this?

Christobal · ‎03-24-2014

Hello Steve, hello Chris, thanks for your answers. Due to weekend I couldn't answer earlier 🙂 It seems that I don't have SAS/STAT installed. I will check whether this is available for me. Your hint Chris seems to be a good solution for me. I will try whether this works fine. Best regards

Christobal · ‎03-21-2014

Hello, thanks for your replay. Unfortunately the reg procedure cannot be found. I get the message "ERROR: Procedure reg not found". It seems, that this procedure is not available in SAS9.1.3? Best regards

Christobal · ‎03-21-2014

Hello all, I have the following problem (SAS 9.1.3 user): I have a table containing values like this: 01/2013 2.200.225 units 02/2013 2.250.000 units 03/2013 2.376.999 units .... .... 02/2014 3.000.100 units I am looking for a sas function that generates a trend/forecast for these values (for e.g.: 12/2014 trend: 3.255.000). While searching the web I found a function called "proc timeseries", but unfortunately this function is not available in SAS 9.1.3. Does anyone know a function or solution for my problem? Thanks in advance and best regards Cristobal

Christobal · ‎09-13-2013

Ok, now I tried this solution: data cpu2; set cpu; by server date; dif = dif(usage); lag = lag(usage); if first.server then call missing(dif,lag); pchg = dif/lag; format pchg percentn10.2; run; This is exactly what I was looking for. Thanks a lot again to all of you for your answers!!!!!!!!!! Best regards Christbal

Christobal · ‎09-13-2013

Hello all, thanks a lot for your answers and sorry for not answering earlier. I will try out your suggestions and give a feedback then. Best regards Christobal

Christobal · ‎09-12-2013

Hello All, I have the following question: I have a table that contains the utilisation of a server per day. This table contains data for several servers: Date Server Usage 04.09.2013 server1 104 05.09.2013 server1 120 06.09.2013 server1 114 07.09.2013 server1 100 08.09.2013 server1 99 09.09.2013 server1 100 10.09.2013 server1 103 11.09.2013 server1 103 12.09.2013 server1 100 04.09.2013 server2 40 05.09.2013 server2 41 06.09.2013 server2 44 07.09.2013 server2 39 08.09.2013 server2 40 09.09.2013 server2 40 10.09.2013 server2 49 11.09.2013 server2 41 12.09.2013 server2 43 Now I want to find out the difference between the utilisation of each server per day in %. See this example: Date Server Usage % 04.09.2013 server1 104 05.09.2013 server1 120 15,38% 06.09.2013 server1 114 -5,00% 07.09.2013 server1 100 -12,28% 08.09.2013 server1 99 -1,00% 09.09.2013 server1 100 1,01% 10.09.2013 server1 103 3,00% 11.09.2013 server1 103 0,00% 12.09.2013 server1 100 -2,91% 04.09.2013 server2 40 05.09.2013 server2 41 2,50% 06.09.2013 server2 44 7,32% 07.09.2013 server2 39 -11,36% 08.09.2013 server2 40 2,56% 09.09.2013 server2 40 0,00% 10.09.2013 server2 49 22,50% 11.09.2013 server2 41 -16,33% 12.09.2013 server2 43 4,88% Is there any Procedure or something else that can help me with this question? Background is, that I want to make an analysis about all the servers that have an increase/decrease in usage of more that 10% compared with the day before. Thanks in advance, best regards Christobal

Christobal · ‎03-01-2013

Hello LeRoy, thank you for your answer, too. Your answer and links etc. seem to be very interesting and useful. I will try to adapt this for my "problem". Perhaps I can provide my colleagues another good solution.

Christobal · ‎03-01-2013

Hello Robert, thanks for your answer. I asked about the colors, because the bars are sorted descending. If a group changes the place with another group, some colleagues said that it's hard to realise this. I also said, that it's not useful to get a chart that looks like a rainbow.... So I think I will use a bar chart that's not sorted descending, because then each bar has always the same place.... But I will try to experiment a little bit with the different solutions...

Christobal · ‎02-28-2013

Hello Reeza, yes, descending bar chart seems to be a good one. Thanks for this hint. Dou you (or anyone else) know if it's possible to give each bar an own pattern? Each bar shows one group ("vbar gruppe"). Now every bar has the same colour although I defined 20 different patterns... If this is possible, is it also possible to match one pattern to one group (the names of the groups are static and always the same). For example I have a group called "TeilA" and the bar for this group should always use pattern3, so that the colour for this group is always the same... While searching the web I unfortunately were not able to find any information that might help me 😕 Another "problem" is, that my legend doesn't appear? This is my current code: ods listing; goptions reset=all; goptions target=WINPRTC device=GIF ; options orientation=landscape; ods pdf style=gdis_logo notoc file="D:\TEST.pdf";; axis1 order=(0 to &maxWertGruppe by 200000) color=black label=(a=90 f='Times New Roman' c=maroon 'gemessen') value=(f='Times New Roman'); axis3 label=none value=(f='Times New Roman' angle=60); axis4 value=none label=none; options nobyline; legend1 label=none position=(middle left) offset=(5,) across=1; pattern1 color='CX999999'; /*BLACK*/ pattern2 color='CX0000FF' ; /*BLUE*/ pattern3 color='CXE8D898' ; /*CREAM*/ pattern4 color='CXFFAA00' ; /*GOLD*/ pattern5 color='CX00FF00' ; /*GREEN*/ pattern6 color='CX8C887A' ; /*LIGHTGRAYISHBROWN*/ pattern7 color='CX2A8307' ; /*OLIVE*/ pattern8 color='CX700000' ; /*MAROON*/ pattern9 color='CXFF8000' ; /*ORANGE*/ pattern10 color='CX703070' ; /*PURPLE*/ pattern11 color='CXE06090' ; /*LILAC*/ pattern12 color='CXFFFF00' ; /*YELLOW*/ pattern13 color='CXE0A860' ; /*TAN*/ pattern14 color='CXFF0055' ; /*SALMON*/ pattern15 color='CX00FFFF' ; /*CYAN*/ pattern16 color='CX4F4F4F' ; /*CHARCOAL*/ pattern17 color='CX142233' ; /*DEEPGREENISHBLUE*/ pattern18 color='CXFF00FF' ; /*MAGENTA*/ pattern19 color='CX8C8385' ; /*REDDISHGRAY*/ pattern20 color='CXB1E599' ; /*LIGHTYELLOWGREEN*/ proc gchart data=work.specintpromonatgruppe;; by monat; vbar gruppe/ sumvar=summevonspecint discrete descending type=sum outside=sum maxis=axis3 raxis=axis1 legend=legend1 ; title2 f='Times New Roman' height=3 pct "Monat: #BYVAL1"; run; quit; ods pdf close; ods listing close; Best regards Christobal

Christobal · ‎02-27-2013

Hello Robert! Thanks for your answer. From my point of view a pie chart was the best way to visualise the data. What kind of chart would you recommend? Regards Christobal

Online Status	Offline
Date Last Visited	‎04-21-2016 08:13 AM

Re: SAS Program does not start when started via bat file

Re: SAS Program does not start when started via bat file

SAS Program does not start when started via bat file

Re: Trend function in SAS

Re: Trend function in SAS

Re: Trend function in SAS

Re: Trend function in SAS

Trend function in SAS

Re: Looking for a Procedure that helps me to compare data

Re: Looking for a Procedure that helps me to compare data

Re: Trend function in SAS

Re: Trend function in SAS

Re: SAS Program does not start when started via bat file

Re: SAS Program does not start when started via bat file

SAS Program does not start when started via bat file

Re: Trend function in SAS

Re: Trend function in SAS

Re: Trend function in SAS

Re: Trend function in SAS

Trend function in SAS

Re: Looking for a Procedure that helps me to compare data

Re: Looking for a Procedure that helps me to compare data

Looking for a Procedure that helps me to compare data

Re: Pie Chart Overlay of values

Re: Pie Chart Overlay of values

Re: Pie Chart Overlay of values

Re: Pie Chart Overlay of values