data have; input VarA $ VarB $ VarC $; cards; A 10:55 4.55 A 10:56 4.56 B 10:57 5.02 A 10:58 4.53 ;

proc sql; select a.vara,b.varb,a.varc, case when a.vara='A' then b.varb else '' end as vard, case when a.vara='B' then b.varb else '' end as vare from have a, have b where a.vara=b.vara and b.varb <= a.varb ;

VarA VarB VarC vard vare ------------------------------------------------ A 10:55 4.55 10:55 A 10:55 4.56 10:55 A 10:56 4.56 10:56 B 10:57 5.02 10:57 A 10:55 4.53 10:55 A 10:56 4.53 10:56 A 10:58 4.53 10:58

As to the event, there was a post a couple days ago for "numbering subgroup". The solution for that can be used.

On Wed, 28 Nov 2007 17:08:44 -0500, Randy <randistan69@HOTMAIL.COM> wrote:

>The example that I am constructing is that of an auction. "A" and "B" are >events. The first event that occurs is an "A" (Event 1) and it occurs at >10:55 and the price (VarC) is at 4:55. The next event is again "A" (Event >2) at 10:56 and VarC is 4.56. At this point in time (10:56) I need to >know all the events that have occured prior to that point in time. So at >each point in time after each event; I want to know the history of all the >the prior events. That is what i mean by stacking the data. > The next event is "B" which occurs at 10:57. At this point I want to >know the history of all the prior "B" events. At that point in time there >is only one event B that has occured. > Event 4 is an "A' and I want to know the history of the all the "A" >events occuring prior to that point in time. And so on. > I hope that this helps. > Randy. > > >On Tue, 27 Nov 2007 03:38:09 -0800, Tree Frog <tree.frog2@HOTMAIL.COM> >wrote: > >>Hi Randy >> >>Please explain the logic of what you're trying to do. Can you give us >>some clues about what the variables actually mean/represent? At the >>moment, it's entirely unclear what you want to do, what you mean by >>'stack the data', and why you have redundant columns in your output. >> >>Have another go at explaining what you're after. >> >>Tree Frog >> >>On Nov 27, 4:22 pm, randista...@HOTMAIL.COM (Randy) wrote: >>> Dear All: >>> >>> My data set is as follows: >>> >>> VarA VarB VarC >>> A 10:55 4.55 >>> A 10:56 4.56 >>> B 10:57 5.02 >>> A 10:58 4.53 >>> >>> I want to stack the data after each "event" that is VarB. So I want >>> >>> VarA VarB VarC VarD VarE Event >>> A 10:55 4:55 4.55 1 >>> A 10:55 4.55 4.55 2 >>> A 10:56 4.56 4.56 2 >>> B 10:57 5.02 5.02 1 >>> A 10:55 4.55 4.55 3 >>> A 10:56 4.56 4.56 3 >>> A 10.58 4.53 4.53 3 >>> >>> Any help will be appreciated. >>> Randy