RE: [amibroker] Backtest start point(s) (was various other subjects)

["Dave Merrill" <dmerrill@xxxxxxx>]

<SPAN 
class=050360921-19102003>isn't it possible that a system that can only have a 
finite number of positions open at once would make an entirely different 
sequence of trades, depending on when you started? the same trades would qualify 
as tradeable, but different ones could be entered. if the day you started, all 
open portfolio positions were filled with trades that held for a long period, 
locking out all other trades, the quality of those specific picks could matter a 
lot.
<SPAN 
class=050360921-19102003> 
<SPAN 
class=050360921-19102003>should it all even out after enough time? I'd think 
that overall performance should average out, since on average the return from 
the trades you entered and didn't enter should average out. but there's no 
inherent reason to think that the trade lists would converge over 
time.
<SPAN 
class=050360921-19102003> 
<SPAN 
class=050360921-19102003>dave
<BLOCKQUOTE 
>
  <FONT face=Tahoma 
  size=2>From: Chuck Rademacher 
  [mailto:chuck_rademacher@xxxxxxxxxx]Sent: Sunday, October 19, 2003 
  3:07 PMTo: amibroker@xxxxxxxxxxxxxxxSubject: [amibroker] 
  Backtest start point(s) (was various other subjects)
  It's 
  quite simple, really.   If you have a system that generates hundreds 
  or thousands of trades and you remove (say) 1% of those trades, the equity 
  curve is unlikely to be affected very much.   As Jitu says, the 
  equity curve quickly "catches up".   If you have a system that 
  doesn't trade very often and you remove 10% of the trades, the resulting 
  equity curve could look quite different.   Also, if Murphy's Law 
  played a hand and the trades you removed were all profitable, the effect on 
  the rest of the equity curve could be 
significant.






Yahoo! Groups Sponsor












Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx 
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html



Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.