|
PureBytes Links
Trading Reference Links
|
> I suppose Mark Brown, alias RD, is laughing all the way to the Houston
> banks.
I think it's Dallas but then all those Texans sound alike. :-)
> I am a bit confused with TS2k and its system testing.
Welcome to the club.
> For example
> when using data2 = advances for the oddball code,
I prefer advances minus declines.
> slightly different
> results
> are obtained if data2 is hidden or visible in subgraph2.
Strange. I don't have a clue about that except maybe the first trade is
on a different date. Maxbarsback on multiple data series is tricky.
> Also, even if
> natural hour bars are used for data1 and data2 it made a difference to put
> a start time and stop time in the code.
I've never gotten the natural hour bars to work right. Mark likes 'em
because that's how the Ned Davis software is set up. The way he gets
them to go in TS is to change the symbol universe for all stock related
symbols to 9am-4:15pm ET. 9am for the natural hour bars and 4:15 to
match the futures pit close. For this particular system, you could plot
30 minute bars and just run the calcs on every other bar.
if mod(time,100) = 0 or time = sess1endtime then begin.....
> Then further optimization on the
> length and the buy level and short level resulted in the two attached
> charts. I really am puzzeled by this.
The main thing about the length and various bar compressions, at least
if you want to stick close to the original, is you are comparing the
breadth today to the breadth yesterday at the same time. So the lookback
is set by the number of bars in a day. You might need a print statement
to make sure you are really looking at the correct bar yesterday.
--
Dennis
To unsubscribe from this group, send an email to:
realtraders-unsubscribe@xxxxxxxxxxxxxxx
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|