|
PureBytes Links
Trading Reference Links
|
Craig DeHaan writes:
> There's a problem loading sufficient days for the HHV and LLV to
> properly evaluate in your setup. Even with 300 days loaded sometimes
> BarsSince(H=HHV(H, 250)) won't get you there, where HHV(H,250) is
> untrue when looking behind the earliest input day period, particularly
> on falling stocks) and the function returns invalid results or N/A.
> Same goes for the LLV, but most frequently in rising stocks where
> there were lower LLVs prior to day -250. It really shouldn't have to
> be this confounding (Equis), but...
"Confounding"... you're being kind. =) I'd call it a bug, plain and
simple. There is no mathematical reason why you can't get the lowest
value in the last 250 days if you *HAVE* all 250 days. But anyway...
I tried it with the default 1300 bars, and same thing there too. Odd,
that.
> Try this as a workaround:
>
> ColA: HHV(H, 250)
> ColB: HHVBars(H, 250)
> ColC: LLV(L, 250)
> ColD: LLVBars(L, 250)
>
> Filter: ColB < 10 and ColD > 240
*DING DING DING DING DING*! That's the ticket. That works fine for me
too.
> Its spins with no N/As this way for me. Oh, and _real_ geek would
> grasp this stuff intuitively, right? I'm so far from it I took a
> couple hours testing just so I could look like one now. =)
Well... I'm not sure how to answer for fear of insulting you or
humiliting me. But, as soon as I read your answer, I realized what you
were doing. I just failed to research enough of MS formula language to
know about LLVBars() and HHVBars(). <egg on face>
Thanks so much; this works great.
========================================================================
Creed of the Engineer: Measure with micrometer, mark with chalk, cut
with axe. (Hope like hell.)
http://www.unixgeek.com/cgi-bin/motd.pl - PGP email preferred
|