About ArtC

ArtC · ‎06-21-2011

Thank you for the confidence Ksharp, but I suspect those may have been someone else's recommendations. Bob Virgle did a fair number of tests on these statements in his efficiencies book a number of years ago. And while I did write about the relative efficiencies in the narrow context of performing table look ups. I would defer to Cynthia's comment that you need to know your data, your machine, your situation. That said I find the differences between the comparisons on numerics and characters to be very interesting and probably worthy of more research.

ArtC · ‎06-20-2011

Try reposting this in the Statistics forum - you may get a better response.

ArtC · ‎06-15-2011

The NOTSORTED option can be used when building a format. This allows you to have the advantage of a format, but maintain the order of the formats definition. This can be very useful when trying to both control order and format a column. It can also avoid the need to create an ordering variable.

ArtC · ‎06-15-2011

I am not a zOS user but would suspect that you need to add: (untested code) [pre] filename outplt 'filelocation.on.your.system...emf'; goptions reset=all; goptions gsfname=outplt device=emf; [/pre]

ArtC · ‎06-15-2011

can we assume that you have experimented with curvelabelposition and curvelabellocation already?

ArtC · ‎06-15-2011

Without seeing more of your code and the data it is a bit hard to be definitive. Perhaps the incoming value is not actually character but is already a formatted datetime value. How have you verified that this is not the case. Are the conversion notes being generated by the statements that you have flagged? All that we can be sure of at this point is that something is not what it seems on first glance.

ArtC · ‎06-15-2011

I cannot address the differences between 9.1.3 and 9.2 for PROC IMPORT, but given that you are getting character values, have you tried using the COMMA. format with an INPUT function?

ArtC · ‎06-14-2011

Here is a PROC REPORT step that has a row counter. [pre] data names; input id name $; datalines; 1 stu 2 Bob 3 Ben run; proc report data=names nowd; columns id name obsno; define id/display; define name/display; define obsno/computed; compute obsno; cnt+1; obsno=cnt; endcomp; run; [/pre]

ArtC · ‎06-14-2011

try the INPUT function. [pre] data dates; cdate='01JUN2011:10:41:27'; datetime = input(cdate,datetime18.); date = datepart(datetime); put cdate=; put datetime= datetime= datetime18.; put date= date= date9.; run; [/pre] the LOG shows [pre]cdate=01JUN2011:10:41:27 datetime=1622544087 datetime=01JUN11:10:41:27 date=18779 date=01JUN2011 [/pre]

ArtC · ‎06-14-2011

Try this. Obviously you never want the last observation. [pre]data ages; input Policy Location Year Age; datalines; 142 12 1978 32 143 05 1980 31 144 17 1990 20 145 05 1992 31 146 17 1995 39 run; data usenext; set ages(firstobs=2 keep=age rename=(age=nextage)); set ages; if nextage>30; run; proc print data=usenext; run; [/pre]

ArtC · ‎06-14-2011

Try posting this to the EG forum.

ArtC · ‎06-13-2011

try adding a ods destination. [pre] ods graphics on; ods pdf body= 'c:\temp\genplot.pdf'; proc genmod data=Crash; class ID year; model accident= major minor/dist=nb link=log offset=wt type3 wald; assess link / crpanel; run; quit; ods pdf close; [/pre]

ArtC · ‎06-13-2011

I have never been very successful with the split= option. However repeating a justification option will cause the text to split. [pre] axis1 value = (t=1 j=c '11' j=c 'years' t=2 j=c '12' j=c 'years'); proc gplot data=sashelp.class; plot weight* age/haxis=axis1; run; quit; [/pre]

ArtC · ‎06-08-2011

In the ODS HTML statement use the NOGTITLE option (to move the title out of the graph). In the TITLE statement you can use the LINK= option or since you are witting in HTML you can use an anchor tag.

ArtC · ‎06-08-2011

You might be able to solve this on the TRANSPOSE side. My guess would be that you did not use the ID statement. the transpose would look something like: [pre] proc transpose data=old out=new; by quarter; id prod; run; [/pre]

Online Status	Offline
Date Last Visited	‎06-20-2021 07:11 PM