Tomorrow I will try to implement it on my case, let's see what happen! Thanks 🙂 ... View more

Thank you for pointing me that out... It is really interesting, but if I understand well, it works just with string based measures (count/count distinct/...) and not with numeric summable ones... ... View more

It would be more somenthing like sum [by event] (QTY) / sum [by district] (POP) * 10000 ... View more

Let's say I have these data: City Dist Event Qty Pop City1 Dist1 Even1 2 100 City1 Dist1 Even2 4 100 City1 Dist1 Even3 6 100 City1 Dist1 Even4 4 100 City1 Dist2 Even1 1 150 City1 Dist1 Even1 2 100 City1 Dist1 Even1 4 100 City2 Dist1 Even4 15 200 City2 Dist1 Even1 7 200 City2 Dist2 Even1 9 250 If I compute the rate by row, I obtain the correct measure. But if I want to aggregate by city, for city1 ie I will obtain Qty=2+4+6+4+2+1+4=23 but I will also obtain Pop=100+100+100+100+150+100+100=750, that's wrong, since the right population is just 100+150=250. ... View more

The problem is not on the per 100.000 part, but for the aggregation... I need the population only aggregate through city and district, but not for eample by event or other variables. If I put the population on the main dataset, I will have pop be summed for each event, giving completly wrong results. On the other hand, if I put the population on another dataset, as far as I know, I will not be able to construct the measure using measures from different data sources ... View more

Ups, never mention... We are on VA 7.5 ... View more

Ok, I think I may not have expressed myself clearly. I'd try to make a better example for my problem...   Suppose you have these two tables of event and population: data TEST; length CITY DISTRICT EVENT $5 QTY 8; input CITY DISTRICT EVENT QTY; datalines; City1 Dist1 Even1 2 City1 Dist1 Even2 4 City1 Dist1 Even3 6 City1 Dist1 Even4 4 City1 Dist2 Even1 1 City1 Dist1 Even1 2 City1 Dist1 Even1 4 City2 Dist1 Even4 15 City2 Dist1 Even1 7 City2 Dist2 Even1 9 ;; run; data POPULATION; length CITY DISTRICT $5 POP 8; input CITY DISTRICT POP; datalines; City1 Dist1 100 City1 Dist2 123 City2 Dist1 45 City2 Dist2 96 ;; run; Now on VA I have a cross table with a City > District hierarchy and a filter on events. What I need is this:   No event filter active:   City Qty Pop Rate City1 23 223 23/223x10000 City2 31 141 31/141x10000   No event filter active but hierarchy open: City District Qty Pop Rate City1 Dist1 22 100 22/100x10000 City1 Dist2 1 123 1/123x10000 City2 Dist1 22 45 22/45x10000 City2 Dist2 9 96 9/96x10000 But if I activate a filter on an event, let's say event1, I need these output: City Qty Pop Rate City1 9 223 9/223x10000 City2 16 141 16/141x10000 City District Qty Pop Rate City1 Dist1 8 100 8/100x10000 City1 Dist2 1 123 1/123x10000 City2 Dist1 7 45 7/45x10000 City2 Dist2 9 96 9/96x10000 So, basically, the Qty depends on the event variable, while the population depends only from City and District variabiles. ... View more

I'm trying to convince customer to update the M1 since almost 4 years...but still here 😞 The code is quite simple proc sort data=sashelp.class out=TEST; by SEX; run; title "Some title referred to M and F"; proc report data=TEST missing; by SEX; columns NAME AGE; define NAME / group; define AGE / mean; run; Same code is used to produce an xml/xls or a pdf, so the solution should be not tightened to the output. For now just write some macro code that get all the by group value and itearate over it... ... View more

Hi, as far as I know, excelxp does not support ods text, isn't it?   Also, a way to force the title output would be good, since I could put the title, force it, then delete the title and make the report... ... View more

Hi, we've tried with both 7.5 and 8.4, but with no luck... In all the version it expect that data source linking is build on columns available in the report object from witch you make the link on the other... At least as far as we have been able to see ... View more

Ok, the result is the same I've would have with the use of the proc report... Sorry for the wrong data, I really need some holidays I suppose... 😓 ... View more

implies either a variable or something with the value G1 in the variable Group but you do not show it anywhere in your data. If AT is supposed to be the totals of the A values, say so, same with the C Sorry, my mistake, forgot to put the gorup in the data, corrected the first post, and added the clarification for the partials AT and CT. ... View more