[Babase] CYCSTATS and REPSTATS Questions - Estimated Matdates
and Gaps
Catherine Markham
amarkham at princeton.edu
Fri Jun 30 14:43:36 EDT 2006
Hi Karl,
I'm posting another follow-up after our phone call this morning. This
one is mostly just for documentation (hopefully I captured the gist of
it all correctly).
Catherine
> 2. Also, PIC seems off. I checked CYCGAPS and she has 2 entries: an E
> on 31 Dec 89 and a P on 1 Jan 90. Her matured date is NOT estimated and
> was on 17 Dec 81 (this date corresponds to a tdate, so that's good). All
> of her cycles are listed as series = 1, despite the CYCGAPS records. I
> think this is because the end date on 31 Dec 89 falls in between a mdate
> and ddate of a single cycles - does that sound right? The mdate of this
> cycle is 24 Nov 89 and the ddate is 1 Jan 90 (this ddate is the same as
> her P day in CYCGAPS). Her CYCSTATS records start on her matured date
> and run through her E day in CYCGAPS - this looks to be continuous (I
> didn't notice and obvious skips).
>
> I think I see what's going on with CYCSTATS and why records stop on the
> E day, but why did MTDCYCLES join that ddate that was a P day in the
> same cycle as the mdate prior to the E of a gap? Do you think it is
> right that they are displayed as a single, linked cycle or should the
> mdate of 24 Nov 89 have no ddate and the ddate on 1 Jan 90 have no mdate
> and be listed as series = 2?
>
> I'm presenting all this without having checked the actual datasheets
> yet. I'll do that later this morning, but felt that this raises an
> important question for how CYCGAPS deals with cycles even if we end up
> getting around the issue by correction cycle dates in this case.
>
> Thanks,
> Catherine
Catherine will look into making sure that PIC's data in BaBase is indeed
accurate with regard to the photocopies. Regardless, these records did
uncover a quirk in how CYCGAPS data is being used to create independent
series in CYCPOINTS. Apparently, the trouble arises with PIC because
the end of a gap is followed the very next day by a point observation
(so those two days are actually "continuous" in the sense that one
follows directly after the other). Karl is going to look into
correcting the system. For PICs records in particular, both Karl and
Catherine agree that (1) the final ddate on 1 Jan 90 should be listed in
a cycle separate from the mdate on 24 Nov 89 and that (2) the cycle
including the 1 Jan 90 ddate should be listed as series "2".
More information about the Babase
mailing list