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

Re: [amibroker] Re: Managing drawdowns (was % channels)



PureBytes Links

Trading Reference Links




Yes, I did notice it. But it was still in the same paragraph where it 
appeared as though you were endorsing martingale betting. I guess I don't 
understand what circumstances would warrant its 'rare' use. What do you mean by 
"...when the nature of the signal warrants it"? In my opinion, one should choose 
a certain MM approach and stick to it. 
 
AV
 
----- Original Message ----- 
<BLOCKQUOTE 
>
  <DIV 
  >From: 
  palsanand 
  
  To: <A title=amibroker@xxxxxxxxxxxxxxx 
  href="">amibroker@xxxxxxxxxxxxxxx 
  Sent: Thursday, October 30, 2003 7:25 
  PM
  Subject: [amibroker] Re: Managing 
  drawdowns (was % channels)
  I agree, that is why I said I use martingale method only 
  very rarely, i.e, only when the nature of the signal warrants it....  
  Maybe you did not notice it or understand the significance of 
  it...rgds, Pal--- In <A 
  href="">amibroker@xxxxxxxxxxxxxxx, "Al 
  Venosa" <advenosa@x...> wrote:> 
  Pal:> > Martingale betting will kill you regardless of your W/L 
  ratio. You mentioned trend-following systems. Such systems notoriously 
  have a winning percentage less than 50%, usually around 40-45%. So, when 
  you have a moderately large losing percentage (55 to 60% or more), the 
  probability of your experiencing 5 or 6 or 7 losses in a row is 
  moderately high (0.6^6 = 4.7%). It makes absolutely no difference what 
  your W/L ratio is. It could be 10:1. But if you double your bet size every 
  time you lose, you lose big time if you don't get that one big 10x win 
  inbetween. Conversely, it makes eminent sense to increase your bet size as 
  you win. After all, you can look at it as betting the market's money 
  rather than your own. If you continue to keep your risk percentage 
  constant but increase the percentage of profits you risk on each trade, 
  your winnings will compound enormously. In other words, risk 1% of your 
  capital on all bets but if your profits increase by, say, 10%, risk 5% of 
  the profits plus your normal 1% of current capital. You'll make big money 
  without increasing your real risk. And if you lose, you LOWER your bet 
  size, not raise it. That way, you stay in the game in case you experience 
  a run of losses. > > Al Venosa>   ----- 
  Original Message ----- >   From: palsanand 
  >   To: amibroker@xxxxxxxxxxxxxxx >   Sent: 
  Thursday, October 30, 2003 6:35 PM>   Subject: [amibroker] 
  Re: Managing drawdowns (was % channels)> > >   
  Thanks for link.  Something interesting mentioned there:> 
  >   All the multitude of money management algorithms may be 
  divided in >   two principal classes: martingale and 
  antimartingale. > >   Martingale methods state that 
  the risk should increase as the capital >   decreases. 
  These methods are popular with traders trying to extract 
  >   profit from a series of losses or break-even 
  trades.> >   Let us review an application of 
  martingale in roulette. We bet 1$ on >   a color and 
  every time we lose, we double the bet. Next time after we 
  >   win, we start at 1$ again. If we lose 10 times in a row, 
  which may >   happen with a probability of (19/37)^10 or 
  0,13%, we'll have to bet >   $1024 to win $1. Since in 
  such a case the expected profit/risk ratio >   is 
  disastrously low, it is often supposed that martingale methods may 
  >   not be used in trading. But, one should keep in mind that 
  in popular >   trend-following methods:> 
  >   1) profits are usually 2-3 times larger than losses 
  > >   2) series of small losses or break-even trades 
  are typically >   interspersed with large profits> 
  >   So martingale methods in our opinion deserve a serious 
  study.> >   I found this to be very true.  
  Typically my trades have either a >   series of small losses 
  or more commonly break-even trades >   interspersed with hugh 
  gains which dwarfs these series of small  >   losses or 
  break-even trades, but I use martingale method only very 
  >   rarely, i.e, only when the nature of the signal warrants 
  it....> >   rgds, Pal>   --- In 
  amibroker@xxxxxxxxxxxxxxx, "Fred" <fctonetti@xxxx> 
  wrote:>   > For those who have an interest in a variety of 
  MM techniques here's >   a >   > 
  short course in a lot of what's out there ...>   > 
  >   > >   <A 
  href="">http://www.tsresearchgroup.com/en/articles/public_20020402010706.php 
  >   > >   > --- In 
  amibroker@xxxxxxxxxxxxxxx, "Fred" <fctonetti@xxxx> 
  wrote:>   > > Nice try ... As far as what trading 
  system he used and whether or >   > not 
  >   > > it was viable, how would I know.  As far 
  as what MM techniques he >   > > used, again how 
  would I know except for the fact that it seems to >   
  > say >   > > one of two things, either his stuff 
  doesn't work as demonstrated >   in >   
  > > his book or he didn't think well enough of it to use it as 
  >   opposed >   > to >   
  > > using something else.  As far as your other reference goes, 
  your >   > > either part of that group in which 
  case you should have >   understood >   > 
  > the reply I made and it should have made some sense or you aren't 
  >   > in >   > > which case you 
  don't have a clue as to what you are referring to.>   
  > > >   > > --- In amibroker@xxxxxxxxxxxxxxx, 
  "MarkF2" <feierstein@xxxx> >   
  wrote:>   > > > "LOL, right ..."?  What 
  simulation software did you use?  No >   > > 
  answer? >   > > > I didn't think I'd get one.  
  You don't know what you're talking >   > > 
  about>   > > > unless you've run extensive MCS on 
  these methods with numerous >   > > 
  systems>   > > > and compared them side by 
  side.  I've tested Van Tharp's stuff >   
  and>   > > > would agree that it works.  But 
  there's stuff in Jones' book >   that>   
  > > > produces better risk/reward curves.  BTW, what he did with 
  his >   own>   > > > account 
  (assuming what you say is true) is irrelevant.  How do 
  >   you>   > > > *know* that he did 
  it "using his own methods?"  Did he have a >   > 
  viable>   > > > trading system to begin with? Did he 
  follow it?  If so, which >   > 
  money>   > > > management method from his book did 
  he use and did he follow >   > that? 
  >   > > > Your humble opinion, lol?  This from 
  the guy who banned Tomasz >   > 
  from>   > > > his group, lol!  Look Fred, if 
  you've got nothing better to do >   > 
  than>   > > > sit there and try to start a flame 
  war, please do it with >   someone>   > 
  > > else.  Because you simply don't know what you're talking 
  about >   on>   > > > 
  this.  Have a great day.>   > > > 
  >   > > > >   > > > --- In 
  amibroker@xxxxxxxxxxxxxxx, "Fred" <fctonetti@xxxx> 
  wrote:>   > > > > LOL, right 
  ...>   > > > > >   > > 
  > > IMHO Van K. Tharp's writings on this topic are more on 
  target.>   > > > > >   > 
  > > > And in all honesty I'm not surprised that Jones traded his 
  own>   > > > account>   > > 
  > > (s) into 90+% DD's using his own mehtods.>   > 
  > > > >   > > > > >   
  > > > > --- In amibroker@xxxxxxxxxxxxxxx, "MarkF2" 
  <feierstein@xxxx> >   > > 
  wrote:>   > > > > > Well you didn't do it 
  correctly.  What simulation software >   
  did>   > > > you >   > > > 
  > use?>   > > > > > >   
  > > > > > --- In amibroker@xxxxxxxxxxxxxxx, "Fred" 
  <fctonetti@xxxx> >   > 
  wrote:>   > > > > > > Yes I have which is 
  why I said I must be one of the >   
  ignorant>   > > > ones.>   > 
  > > > > > >   > > > > > > 
  --- In amibroker@xxxxxxxxxxxxxxx, "MarkF2" >   > 
  <feierstein@xxxx> >   > > > > 
  wrote:>   > > > > > > > Have you read 
  and tested the stuff in his book, or are >   
  you>   > > > just >   > > 
  > > > > making>   > > > > > > 
  > a typically provocative comment?>   > > > > 
  > > > >   > > > > > > > --- In 
  amibroker@xxxxxxxxxxxxxxx, "Fred" >   
  <fctonetti@xxxx>>   > > > 
  wrote:>   > > > > > > > > Ryan Jones 
  ? OMG ... I must be one of the ignorant >   
  ones.>   > > > > > > > > 
  >   > > > > > > > > --- In 
  amibroker@xxxxxxxxxxxxxxx, "MarkF2">   > > > 
  <feierstein@xxxx> >   > > > > > > 
  wrote:>   > > > > > > > > > Hi 
  Leo!>   > > > > > > > > > 
  >   > > > > > > > > > Let me 
  elaborate.  Although I wouldn't put $.02 on >   a 
  >   > > > > *simple*>   > 
  > > > > > > > > Martingale or anti-Martingale method 
  of money >   > > management,>   > 
  > > I >   > > > > do >   
  > > > > > > think>   > > > > 
  > > > > > that the latter is certainly viable while the 
  >   former >   > is >   
  > > > > not. >   > > > > > > 
  How to >   > > > > > > > > 
  do>   > > > > > > > > > 
  better?  I'd recommend reading The Trading Game by 
  >   > Ryan >   > > > > Jones 
  >   > > > > > > *and>   
  > > > > > > > > > then running simulations* of the 
  tradeoff between >   > > equity 
  >   > > > > growth >   > > 
  > > > > and>   > > > > > > > 
  > > drawdown for the various methods *for your trading 
  >   > > > > systems*.  I>   
  > > > > > > > > > developed my personal favorites 
  after reading this >   > book>   > 
  > > but >   > > > > > > 
  everyone>   > > > > > > > > > 
  needs to look at their own curves from their own >   > 
  > > > simulations for>   > > > > > 
  > > > > themselves to see what suits them best.  This is 
  a>   > > > tedious >   > > 
  > > > > project >   > > > > > > 
  > > and>   > > > > > > > > > 
  not much fun, but well worth the effort in my >   > 
  opinion.  >   > > > > BTW, if 
  >   > > > > > > you>   
  > > > > > > > > > look at the reviews of this book 
  on amazon, there >   are>   > > 
  > some >   > > > > > > > > 
  *incredibly>   > > > > > > > > > 
  ignorant* ones by people who obviously didn't take >   
  the>   > > > time >   > > 
  > > to >   > > > > > > dig 
  >   > > > > > > > > 
  in>   > > > > > > > > > to the 
  material and do their homework which to me, >   is 
  >   > > > > running>   > > 
  > > > > > > > simulations on all of the methods.  I 
  have and >   trust >   > > me, 
  >   > > > > lol, >   > > 
  > > > > > > there's>   > > > > 
  > > > > > good stuff in this book.>   > 
  > > > > > > > > >   > > > 
  > > > > > > Best Regards,>   > > > 
  > > > > > > >   > > > > > 
  > > > > Mark>   > > > > > > 
  > > > >   > > > > > > > > 
  > --- In amibroker@xxxxxxxxxxxxxxx, "leonardot19" >   > 
  > > > > > > > 
  <leo.timmermans@xxxx>>   > > > > > > 
  > > > wrote:>   > > > > > > > 
  > > > Hi Mark,>   > > > > > > > 
  > > > >   > > > > > > > > 
  > > Which MM technique would you use than, can you 
  >   give >   > > an 
  >   > > > > example>   > > 
  > > > > > > > > please ?>   > > 
  > > > > > > > > >   > > > 
  > > > > > > > Kind regards>   > > 
  > > > > > > > > Leo>   > > > 
  > > > > > > > >   > > > > 
  > > > > > > >   > > > > > 
  > > > > > --- In amibroker@xxxxxxxxxxxxxxx, "MarkF2" 
  >   > > > > <feierstein@xxxx> 
  >   > > > > > > > > 
  wrote:>   > > > > > > > > > > 
  > Neither of these is a technique I'd put $.02 on,>   
  > > > quite >   > > > > > > 
  easily>   > > > > > > > > > > 
  > demonstrated by bootstrapping representative >   
  > trades >   > > > > while 
  >   > > > > > > > > 
  applying>   > > > > > > > > > > 
  > them.  Every time I mention simulation >   
  everyones'>   > > > eyes >   > 
  > > > glaze>   > > > > > > > 
  > > over, >   > > > > > > > > 
  > > but>   > > > > > > > > > 
  > > if you're not using it for position sizing or 
  >   > money >   > > > > > 
  > management >   > > > > > > > > 
  or>   > > > > > > > > > > > 
  whatever you want to call it, you're flying >   
  blind.>   > > > > > > > > > > 
  > >   > > > > > > > > > > 
  > --- In amibroker@xxxxxxxxxxxxxxx, "palsanand" >   
  > > > > > > <palsanand@xxxx> >   > 
  > > > > > > > > > wrote:>   > 
  > > > > > > > > > > > 
  Dave,>   > > > > > > > > > > 
  > > >   > > > > > > > > > 
  > > > There is a good link I came across:>   > 
  > > > > > > > > > > > >   
  > > > > > > > > > > > > <A 
  href="">http://www.arbtrading.com/moneymanagement.htm>   
  > > > > > > > > > > > > 
  >   > > > > > > > > > > > 
  > I like the Anti-Martingale and Martingale >   > > 
  (doubling >   > > > > up) >   
  > > > > > > > > systems >   > > 
  > > > > > > > > to >   > > > 
  > > > > > > > > > manage drawdowns.  I would 
  use a combination >   of>   > > 
  > these >   > > > > > > 
  systems,>   > > > > > > > > > so 
  >   > > > > > > > > > > > 
  > that when I'm losing money I would use >   > 
  Martingale >   > > > > system 
  >   > > > > > > and>   
  > > > > > > > > > when>   > 
  > > > > > > > > > > I'm >   
  > > > > > > > > > > > > finally making 
  money with the final position, >   I>   
  > > > would >   > > > > be 
  >   > > > > > > > > > > > 
  > automatically switched over to Anti->   Martingale 
  >   > > > > system, >   > 
  > > > > > but >   > > > > > 
  > > > may >   > > > > > > > 
  > > > most >   > > > > > > > 
  > > > > > likely exit losing positions at break-even 
  >   > price. >   > > > I 
  >   > > > > would>   > > 
  > > > > > > > double>   > > > 
  > > > > > > > > up >   > > > 
  > > > > > > > > > only when I get stronger signals 
  verfied by >   > OB/OS >   > > 
  > > > > conditions >   > > > > > 
  > > > in >   > > > > > > > > 
  > > the >   > > > > > > > > 
  > > > > subsequent session, so that my system of 
  using>   > > > 3BSMA >   > 
  > > > for >   > > > > > > 
  the>   > > > > > > > > > next 
  >   > > > > > > > > > > > 
  > session is temporarily suspended.  It does >   
  take >   > > > > usually >   
  > > > > > > about >   > > > > 
  > > > > 3>   > > > > > > > 
  > > > > days >   > > > > > > 
  > > > > > > for a trend-change to fully develop.  I 
  would >   > not >   > > > 
  > double >   > > > > > > 
  up>   > > > > > > > > > 
  beyond>   > > > > > > > > > > 
  > 3 >   > > > > > > > > > > 
  > > consecutive days, because if you are wrong 4 
  >   > times>   > > > in 
  >   > > > > a >   > > > 
  > > > row,>   > > > > > > > 
  > > most >   > > > > > > > > 
  > > > > likely the market is starting a new trend in 
  >   > the >   > > > > 
  opposite >   > > > > > > > > > 
  > direction >   > > > > > > > > 
  > > > > and will go against you and so better to 
  >   exit.  >   > I 
  >   > > > > have >   > > 
  > > > > done >   > > > > > > 
  > > this>   > > > > > > > > 
  > > > many >   > > > > > > > 
  > > > > > times, as I find it impossible to optimize my 
  >   > > entry >   > > > > 
  > > points.  >   > > > > > > 
  > > But>   > > > > > > > > > 
  > > the >   > > > > > > > > 
  > > > > safest course is to wait for the 
  actual>   > > > Trend-change >   
  > > > > > > signal>   > > > > 
  > > > > > > > verified >   > > 
  > > > > > > > > > > by OB/OS conditions, then 
  you may never have >   to >   > > 
  > > double up >   > > > > > > 
  but>   > > > > > > > > > 
  you>   > > > > > > > > > > > 
  may >   > > > > > > > > > > 
  > > miss some signals.  This may sound crazy for 
  >   > some>   > > > but 
  >   > > > > it >   > > 
  > > > > does>   > > > > > > 
  > > > seem>   > > > > > > > 
  > > > > to >   > > > > > > > 
  > > > > > work for me especially with the AFL pivot 
  >   > points >   > > to 
  >   > > > > > > predict 
  >   > > > > > > > > 
  the>   > > > > > > > > > > > 
  Next >   > > > > > > > > > > 
  > > bar approximate High/Low of Day and >   appropriate 
  >   > > > > position >   > 
  > > > > > > > sizing.>   > > > 
  > > > > > > > > > >   > > 
  > > > > > > > > > > Regarding whether your 
  system has stopped >   > working>   > 
  > > or >   > > > > not, 
  >   > > > > > > it >   
  > > > > > > > > is>   > > > 
  > > > > > > > > hard >   > > 
  > > > > > > > > > > to say.  I would try to 
  improve the system >   > > > > performance 
  >   > > > > > > using a>   
  > > > > > > > > > > > system 
  >   > > > > > > > > > > > 
  > of filters, stops and walkforward testing.  >   
  > > Easier >   > > > > said 
  >   > > > > > > than >   
  > > > > > > > > > > done...>   
  > > > > > > > > > > > > 
  >   > > > > > > > > > > > 
  > Regards,>   > > > > > > > > > 
  > > > >   > > > > > > > > 
  > > > > Pal>   > > > > > > > 
  > > > > > >   > > > > > > 
  > > > > > > >   > > > > > 
  > > > > > > > --- In amibroker@xxxxxxxxxxxxxxx, "Dave 
  >   Merrill">   > > > > > 
  > > > > <dmerrill@xxxx> >   > > > 
  > > > > > > > > > wrote:>   > 
  > > > > > > > > > > > > I've been 
  wondering, could I trade a system >   > > with 
  >   > > > > 50% >   > > 
  > > > > > > average >   > > > > 
  > > > > > > gain >   > > > > 
  > > > > > > > > per year>   > > 
  > > > > > > > > > > > since '95, and max 
  system drawdown of 40->   50%.>   > 
  > > even >   > > > > if 
  >   > > > > > > I've>   
  > > > > > > > > > seen >   > 
  > > > > > > > > > > > that 
  in>   > > > > > > > > > > > 
  > > backtests beforehand, could I really look >   
  at >   > > that >   > > > 
  > kind >   > > > > > > 
  of>   > > > > > > > > > drop 
  >   > > > > > > > > > > in 
  >   > > > > > > > > > > > 
  > my account>   > > > > > > > > 
  > > > > > and still believe I was doing the right 
  >   > thing? >   > > or 
  >   > > > > would >   > > 
  > > > > I>   > > > > > > > 
  > > think >   > > > > > > > > 
  > > > > it'd finally>   > > > > > 
  > > > > > > > > just stopped working? and if I am able 
  to >   > > ignore >   > > 
  > > that >   > > > > > > much 
  >   > > > > > > > > > > > 
  > drawdown, how>   > > > > > > > > 
  > > > > > would I know if it really *had* stopped 
  >   > working?>   > > > > 
  > > > > > > > > > >   > > 
  > > > > > > > > > > > by the 
  half-the-gain-twice-the-drawdown >   > > > > 
  tolerability >   > > > > > > 
  rule,>   > > > > > > > > > 
  this>   > > > > > > > > > > 
  > is a>   > > > > > > > > > 
  > > > > non-starter.>   > > > > > 
  > > > > > > > > >   > > > 
  > > > > > > > > > > dave>   
  > > > > > > > > > > > > >   
  Defense ... Yep or as I've said it's not >   > 
  what>   > > > you >   > > 
  > > > > make, >   > > > > > > 
  > > it's>   > > > > > > > > 
  > > > what >   > > > > > > > 
  > > > > > you>   > > > > > > 
  > > > > > > >   keep.  DD's are killers 
  from lots of >   aspects>   > > > 
  not >   > > > > just >   > 
  > > > > > in>   > > > > > > 
  > > > terms>   > > > > > > > 
  > > > > of>   > > > > > > > 
  > > > > > >   what they do to your account balance 
  but >   > also >   > > > 
  > what >   > > > > > > they 
  do>   > > > > > > > > > 
  to>   > > > > > > > > > > > 
  ones>   > > > > > > > > > > 
  > > >   ability psycologically to trade and stay 
  >   > with >   > > > > 
  systems >   > > > > > > 
  that>   > > > > > > > > > do 
  >   > > > > > > > > > > > 
  > work.> > 
  >         Yahoo! Groups Sponsor 
  >               
  ADVERTISEMENT>                 
  >        
  >        > > 
  >   Send BUG REPORTS to bugs@xxxx>   Send 
  SUGGESTIONS to suggest@xxxx>   
  ----------------------------------------->   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 
  Yahoo! Terms of Service. > > >   
  --->   Outgoing mail is certified Virus 
  Free.>   Checked by AVG anti-virus system (<A 
  href="">http://www.grisoft.com).>   
  Version: 6.0.525 / Virus Database: 322 - Release Date: 
  10/9/2003
  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. 
  
<BLOCKQUOTE 
><FONT 
  face="Courier New">---Outgoing mail is certified Virus 
  Free.Checked by AVG anti-virus system (<A 
  href="">http://www.grisoft.com).Version: 6.0.525 
  / Virus Database: 322 - Release Date: 
10/9/2003






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.