I'm doing a scatter plot with only two discrete categories on the X axis.  I would like to use two different symbols for each category.  I'm used to using a different symbol per  third grouping, such as sex, but how can I use a different symbol per category?  It seems too simple to do, ha ha.   Here's what I want it to look like:     ... View more

I've recently been creating graphics in PDF output using ODS Layout statements to arrange and size multiple plots on a page.   It seems that this can't be done so easily when writing to RTF.  I would like to confirm (because maybe I am coming across ancient commentary on the internet) that Absolute ODS Layout, and in fact many of the features I used when writing to PDF, are not supported/available when writing to RTF.  Am I correct about that?   Thank you... ... View more

Greetings,   I am using PROC LIFETEST to calculate values for a KM plot, and then using  PROC SGPLOT to plot the curves.  Is there a way to use PDS OUTPUT to capture the CI band values that are shown in this plot that is output directly from LIFETEST?   Here is my code (as you can see, I've been trying to see what output dataset they might be routed to):   ods output CensoredSummary = cs Means = mean quartiles = quart; ods trace on; proc lifetest data = acmort0 alphaqt = 0.05 plots = survival (cb atrisk = 0 to 30 by 3); time aval * cnsr(1); strata cvhgrpn / notest; run; Here is the plot created by LIFETEST, which includes the confidence bands:     Is there any way to capture the band values so I can plot them myself via SGPLOT?   I realize that it's possible to customize the plot created by PROC LIFETEST (a "Warren Kuhfeld special"), but the program I inherited doesn't do it that way and I don't want to disturb it at this point in our project timeline.   Thank you!   ... View more

Somebody wants standard error bars added to a histogram.  I am hoping just to modify my PROC SGPLOT to add.   I thought it would be simple -- just add yerrorlower= yerrorupper = after the slash on the VBAR statement.  But apparently VBAR doesn't like those.   Then I tried a separate SCATTER statement with the error bars.  But again, I got this red error message: ERROR: Attempting to overlay incompatible plot or chart types.   So, can I do a histogram with these bars?  I see pictures of them, so I know the idea isn't insane:   Any ideas?  Thank you.  Here is my code (with the failed SCATTER statement left in. prior to that I tried yerrorlower= yerrowupper= after the slash on the VBAR statement).   proc sgplot data = adcv4 noautolegend; where paramcd = 'LVMASSIX'; vbar tx / group = tx datalabel = txtext datalabelpos=top datalabelattrs = (color=black family="courier new" size=11 weight=bold) filltype=solid outlineattrs=(color='black') response=meanchg; scatter x=tx y=meanchg / yerrorlower=lower yerrorupper=upper; styleattrs datacolors=(viypk vligb); xaxis label = ' ' display = (noticks novalues) offsetmin = .3 offsetmax = .3; yaxis label = 'Mean Percent' labelattrs = (family = "courier new" size = 10 pt color = black weight = bold) values = (-2 to 8 by 2); run;   ... View more

Thank you in advance.   I am trying to force the items in my scatter plot to legend to appear in a certain order (by numeric visit number, but labeled via a format).  So far I can only get them to appear alphabetically, for some reason.  (Also, not sure why SAS is using the same symbol for multiple visits).   Here is my code:   proc format; value visf 1 = 'Baseline' 28 = 'Day 28' 91 = 'Month 3' 183 = 'Month 6' 274 = 'Month 9' 365 = 'Month 12' 457 = 'Month 15' 548 = 'Month 18' 639 = 'Month 21' 731 = 'Month 24' 822 = 'Month 27' 913 = 'Month 30'; run; proc sgplot data = ttrpc0; scatter x = pc y = ttr / name = "scatter1" group = avisitn groupdisplay = cluster markerattrs=(color=black); xaxis label = 'Plasma AG10 Concentration (ng/mL)' labelattrs = (family = "courier new" size = 9.9 pt color = black weight = bold) offsetmin = .04 offsetmax = .04; yaxis label = 'Serum TTR (mg/dL)' labelattrs = (family = "courier new" size = 9.9 pt color = black weight = bold); format avisitn visf. ; run; And here is what the plot currently looks like (a bit busy, but this is what "they" want):     Any ideas from you experts?   If nothing else I thought I could create a custom legend, but before that I would need to resolve the distinct-symbol-per-visit issue.  And I don't feel like that should be necessary; I'm not really doing anything sophisticated here.   Thank you very much... ... View more