[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: whats this FEATURE ?



PureBytes Links

Trading Reference Links


Gary wrote:
> If the exit happens on a bar AFTER the maximum high/low was hit, then the
> internal trailing stops are (as far as I know anyway) safe.

maybe - but what means SAFE: that means you have to INSPECT all your 
charts for the backtest when, where (why<g>) it happened.

I would prefer to RUN THEM and take the results - PERIOD :-)

PO sent the following privately after I forwarded the whole thread to him - as 
that may shed new light I answer publically:

PO ecrit:
> A few questions:
> 
>  What was the exact day where the signal appeared and the day where it
>  disappeared ?

It should be clear by my list mails but here it is again: 
TUESDAY 19th appeared
WEDNESDAY 20th was gone

> Can you send me the daily datafile for the SP symbol

No need besides: 
OHLC 127700 128900 126650 127500, i.e. it closed lower than it opened
 
> Have you a sig5.dat and sig5.idx file ( or maybe signal5.dat) in the omega
> /sercver/ sig directory ?

sure I have - but dont see any need for it - the problem is BOUNCING TICKS -
the very worst of it that TS has changed a REALTIME trade with hindsight 
after the EOD close !

As I said it was set at 10%, changed it as good ole Dennis suggested aka 
workaround (tmŽ ORŽ) to 100% and the EXIT is BACK again were I saw it.

BTW: the system as per FAKE_TS  (tmŽ ZEBRAŽ) still shows in the old 
entry in the system with BT at 10% set as open trade and has a 23.0 bps 
open profit - nice make up for a trade which already was bailed in REALTIME 
at 16 BPS LOSS !!!!!!!!!!

> In the Server option have you checked the morning update ?

yes, obviously it gets triggered there - so with UNchecking this, it MIGHT not 
do the FAKE_UPDATE - still the BT is a major problem for any backtest

thanks for any insight - especially whether setting the BT to 100% is a SAFE 
guarantee, which I just dont believe......

I just did a test of the same system on S&P on EOD data with BT set at 0% 
and BT set at 100% - see attached GIF....... its a HUGHE difference if you 
watch the ROA !!! - and obviously it would be even more different running in 
REALTIME and the worst: I have no clue whether better or worse !!!

rgds hans