Date:         Wed, 19 May 2004 13:49:57 -0700
Reply-To:     cassell.david@EPAMAIL.EPA.GOV
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "David L. Cassell" <cassell.david@EPAMAIL.EPA.GOV>
Subject:      Re: Choosing an episode
Comments: To: "Beckman, Anthony" <Anthony_Beckman@URMC.ROCHESTER.EDU>
Content-type: text/plain; charset=US-ASCII

"Beckman, Anthony" <Anthony_Beckman@URMC.ROCHESTER.EDU> replied:
> I wish to have a value from the epoch before the first zero rating and
a
> value from the epoch after the zero not the rating at the end of a
run.

Okay, then I still would like some clarification about how you wish
to interpret your data.  What about:

        epoch   rating
        1       3
        2       3
        3       2
        4       0
        5       0
        6       0
        7       0
        8       2
        9       5
        10      0
        11      3
        12      0
        13      4
        14      0
        15      2

You seem to want an interval to go from #3 to #5.  Are you then going to
'lose'
the zeroes at #5 and #6 and #7 ?

You want the next interval to go from #9 to #11 .  Okay, do we re-set
instantaneously, so the next interval goes from #11 to #13, or do we
lose
that and start over at the next postivie-number-followed-by-zero
sequence?
If so, then why aren't we re-setting instantaneously within the #3 to #8
sequence?  Perhaps the list simply hasn't been given the underlying
systemata
to clarify what you are really trying to achieve.

Could you write back to the list (not to me personally) and clarify, so
that we might help you more?

HTH,
David
--
David Cassell, CSC
Cassell.David@epa.gov
Senior computing specialist
mathematical statistician