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

RE: [amibroker] detecting "basket" performance in a backtest



PureBytes Links

Trading Reference Links




<SPAN 
class=311370119-29082003>I'm coming from a VectorVest world, and like many, I 
think VV picks stocks well given good timing, but you need to find a timing 
signal that works well. once you have that, the idea is to be long or short 
always, depending on the state of your market indicator, using some long or 
short VV strategy to pick the stocks to hold. any indicator you use will 
sometimes be "wrong", meaning that shortly thereafter the market didn't do what 
the indicator indicated. the idea behind the basket stop is to use the overall 
performance of the of group stocks you bought at that time to validate the 
"truth" of the overall market signal, and sell everything if the basket overall 
drops some amount.
<SPAN 
class=311370119-29082003> 
<SPAN 
class=311370119-29082003>but with all due respect, I wasn't asking for an 
evaluation of the idea -- the backtester will do that just fine 
(:-)
<SPAN 
class=311370119-29082003> 
<SPAN 
class=311370119-29082003>...if I can figure out how to backtest 
it!
<SPAN 
class=311370119-29082003> 
<SPAN 
class=311370119-29082003>the thing I can't figure out is how to access the list 
of stocks currently owned during a backtest.
<SPAN 
class=311370119-29082003> 
<SPAN 
class=311370119-29082003>anyone?
<SPAN 
class=311370119-29082003> 
<SPAN 
class=311370119-29082003>dave
<BLOCKQUOTE 
>
  Buying and selling 
  the whole basket depending on a market signal will work only if your 
  basket matches the market (however you defined the market).  If you use a 
  market index like the DJ or S&P your basket must closely follow these 
  indices.  If you are concerned about a basket stop an easy way would be 
  to buy an extraded fund that follows a 
  major index.
  <FONT face=Arial 
  size=2> 
  There are times 
  when some major indices are down and most stocks are up and 
  vice versa. 
   
  The idea of a 
  basket stop sounds appealing, but unless you have a large number of stocks in 
  your basket why sell them all unless you think that the market is going to 
  drop precipitously? With 10 to 20 stocks in your basket, you should 
  be able to track them with trailing stops.
  <FONT face=Arial 
  size=2> 
  However if this 
  idea works for you, go ahead and use it.
   
  <FONT face=Arial 
  size=2>Lionel
  
  
  <FONT 
  face=Tahoma size=2>-----Original Message-----From: Dave Merrill 
  [mailto:dmerrill@xxxxxxx] Sent: Friday, August 29, 2003 1:17 
  PMTo: amibroker@xxxxxxxxxxxxxxxSubject: RE: [amibroker] 
  detecting "basket" performance in a backtest
  <SPAN 
  class=085511118-29082003>the thing I was wondering about was the notion of a 
  basket stop: interpreting an overall drop in the basket purchased at one time 
  as a buy signal error, and selling them all. this is in contrast to stops set 
  on individual stocks.
  <SPAN 
  class=085511118-29082003> 
  <SPAN 
  class=085511118-29082003>dave
  <BLOCKQUOTE 
  >
    10% was 
    suggested by Curtis Dahl when he published his book back around 1960, 
    and Nicholas Darvas  used 10% stops.  At that time, before 
    the days of the PC, markets were much less volatile and the trader had more 
    time to make a decision. More recently some people have told me numbers 
    between 3% and 5%.  
    My own feeling 
    is that if any of your stocks are down 10% sell them, or if you can't 
    bear to give a sell order put in stop-loss orders.
    <FONT face=Arial 
    size=2> 
    I don't think 
    that you need to do a formal back test to verify this
    <FONT face=Arial 
    size=2> 
    <FONT face=Arial 
    size=2>Lionel
     
     when using an overall 
    market timing signal plus some strategy to pickspecific stocks, someone 
    on another list proposed that if the overallperformance of all stocks 
    purchased at that time dropped by some amount (10%say), the whole basket 
    should be dumped. the theory is that the overallmarket signal was 
    "wrong".is there any way to backtest this kind of behavior in AB? 
    how can I accessthe value of all stocks currently 
    held?thanks,Dave MerrillSend BUG 
  REPORTS to bugs@xxxxxxxxxxxxxSend SUGGESTIONS to 
  suggest@xxxxxxxxxxxxx-----------------------------------------Post 
  AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx (Web page: <A 
  href="">http://groups.yahoo.com/group/amiquote/messages/)--------------------------------------------Check 
  group FAQ at: <A 
  href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html 
  Your use of Yahoo! Groups is subject to the <A 
  href="">Yahoo! Terms of Service. 
  Send 
  BUG REPORTS to bugs@xxxxxxxxxxxxxSend SUGGESTIONS to 
  suggest@xxxxxxxxxxxxx-----------------------------------------Post 
  AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx (Web page: <A 
  href="">http://groups.yahoo.com/group/amiquote/messages/)--------------------------------------------Check 
  group FAQ at: <A 
  href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html 
  Your use of Yahoo! Groups is subject to the <A 
  href="">Yahoo! Terms of Service. 







Yahoo! Groups Sponsor


  ADVERTISEMENT 









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.