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

[amibroker] Re: FW: Optimize/OverOptimize



PureBytes Links

Trading Reference Links

Steve,

At first, I was puzzled why you attacked me like that, but only until 
I re-read Dave's original post. I missed the part of that post 
explaining that the views discussed were presented by you in one of 
your seminars, and thus understand why one would take my post in a 
rather personal manner. Whether you did or not, does not matter. What 
I wrote, and although I stand by my beliefs, was not appropriate as a 
response to Dave's post. I was trying to present my views about 
mistakes made by novice traders when it comes to optimization and/or 
believing everything written in myriad of books available on trading.

Anyway, please accept my apologies (and apologies to Dave too, for 
misunderstanding what he wanted to discuss). I meant no disrespect to 
you or anyone else. That was not the intention.

As to your observation about my inability about understanding of back 
and forward testing, I respectfully disagree, and would in fact love 
to debate the pros and cons, but this thread is not appropriate for 
that. Perhaps some other time.

Jitu

--- In amibroker@xxxxxxxxxxxxxxx, "CedarCreekTrading" <kernish@xxxx> 
wrote:
> May be I'll learn something.
> 
> jtelang,
> 
> I doubt it...but, pay attention.  I don't believe you even have a 
grasp of how the credible people approach their research (of which 
back and forward testing is an important part).
> 
> Take care,
> 
> Steve
>   ----- Original Message ----- 
>   From: jtelang 
>   To: amibroker@xxxxxxxxxxxxxxx 
>   Sent: Wednesday, October 08, 2003 8:03 PM
>   Subject: [amibroker] Re: FW: Optimize/OverOptimize
> 
> 
>   Dave, if this thread goes on for ever, YOU will be the one to 
>   blame. :-) Here! I'll help you completely open the can of worms 
that 
>   you're attempting to open. :-P
> 
>   IMO, most, but not all, attempts to optimize are based on flawed 
>   logic. I'll present myself as an example, which should rhyme with 
>   most novice traders. A typical wanna-be trader reads Dr. Elder's 
>   book. Then he/she tries to implement it. Of course, Dr. Elder, 
like 
>   many other "experts", never offers a completely working system, 
>   although the book offers, like many other books, good concepts 
well 
>   understood by practically reading any other trading related book 
if 
>   you have even half the brains of what it'd take for you to 
survive in 
>   this jungle. Instead, just concepts. Now here, substitute YOUR 
>   favorite book for Dr. Elder's book. Same thing applies.
> 
>   So its upto me, an average person, to derive a profitable system 
from 
>   his concepts. Hmmm... What are the odds of that happening? After 
all, 
>   HE had to resort to selling stuff to other traders to make a 
living 
>   himself. Call me fanatically skeptical, but I find it funny how 
that 
>   works.
> 
>   Now, you are thinking... What does this have to do with 
optimization? 
>   It does, a lot. Because optimization, IMHO, is an attempt to make 
an  
>   unproven logic work with the past data you have. Read that 
definition 
>   again. That's all it is. At least, its a good thing that the base 
>   concept is not "completely" random, if you read a book or two. 
But 
>   nonetheless, its true. You start with a concept that you "think" 
will 
>   work, then it doesn't, then you tweak it a little bit, it still 
>   doesn't work, and then you finally turn to the software to tweak 
it 
>   to death to make it work with the past data, a la optimization.
> 
>   Think about it this way, if someone could become the greatest 
stock 
>   trader just by optimizing a concept on the past data, wouldn't we 
>   have heard about him/her by this time? That itself should answer 
your 
>   original question.
> 
>   IMHO, one should have a "sound" strategy in the first place. A 
sound 
>   strategy should take into account ALL factors that the great 
traders 
>   of the past have known to take into consideration. Bounce it 
around a 
>   few people, if you're not sure that its sound. And THEN you can 
>   optimize it a LITTLE BIT, and THAT is ok.
> 
>   All of this of course, is IMHO, a novice trader. I'm all for 
getting 
>   flamed by someone who is actually making tons of money just by  
>   optimizing an original strategy that didn't work prior to the 
>   optimization. May be I'll learn something.
> 
>   And if not ualready nderstood, no offense meant to ANYONE. 
Neither 
>   people who optimize, nor people who are fans of Dr. Elder. Direct 
all 
>   flames at Dave. :-)
> 
>   Jitu
> 
>   PS: A family emergency will prevent me from replying to this for 
>   couple of days, but I will follow it up as soon as I get a chance.
> 
>   --- In amibroker@xxxxxxxxxxxxxxx, "Dave Merrill" <dmerrill@xxxx> 
>   wrote:
>   > [other reply to my questions, from Dave Chamness. - dave 
merrill]
>   > 
>   > -----Original Message-----
>   > From: David Chamness
>   > Subject: Re: Optimize/OverOptimize
>   > 
>   > 
>   > It's OK to post my replies.
>   > 
>   > Equities may trade based on a specialist, or a small group of 
>   frequent
>   > traders, in effect market makers.  Change the group and price 
>   behavior may
>   > change.  Try to find a system for GE.  You may have better luck 
>   with a
>   > smaller stock.
>   > 
>   > Commodities are traded by many people who do not need to make a 
>   profit, such
>   > as hedgers and governments.  Currencies and interest rates 
trend 
>   because
>   > Alan Greenspan does not want to look like an idiot jacking 
rates up 
>   and down
>   > in a random walk.  So he lowers interest rates repeatedly until 
he 
>   is done.
>   > 
>   > Personally, I trade commodities, but I keep searching for stock 
>   systems.
>   > 
>   > Dave
>   >   ----- Original Message -----
>   >   From:  dave merrill
>   >   Subject: RE: Optimize/OverOptimize
>   > 
>   > 
>   >   thanks for clarifying, much appreciated. is it ok w you if I 
>   forward your
>   > reply(ies)to the AmiBroker group where steve posted your 
original? 
>   let me
>   > know.
>   > 
>   >   one area intrigues me still:
>   > 
>   >   if we do find a market where a simple rule set works well, 
why 
>   would you
>   > think that's so? because of some inherent property of the stock 
>   itself that
>   > makes it non-random, different from other issues where that 
rule 
>   fails? or
>   > is it another random walk phenomenon, unlikely to persist at 
all? 
>   if that's
>   > so, it seems completely pointless to trade equities at all, no 
>   different
>   > from gambling.
>   > 
>   >   why do you think commodities act differently? because prices 
>   respond more
>   > to real-world changes (supply/demand and factors that influence 
it, 
>   etc)
>   > than to the raw emotionality that seems to drive equities? if 
so, 
>   that
>   > implies we should look to fundamentals for more non-random 
trends in
>   > equities, but not much in that dimension except news spikes 
seems 
>   to drive
>   > valuation very much. how do we resolve this apparent lack of 
>   perceivable
>   > order, other than trading commodities instead?
>   > 
>   >   thanks again for your thoughts, very interesting.
>   > 
>   >   Dave Merrill
>   > 
>   >     Answers are in the text below.  Contrary to Steve's 
statement, 
>   I have
>   > only one degree, BS Mechanical Engineering.
>   > 
>   >     Dave Chamness
>   > 
>   >     -----Original Message-----
>   >     From: Dave Merrill [mailto:dmerrill@x...]
>   >     Sent: Monday, September 29, 2003 12:32 PM
>   >     To: dec@xxxx
>   >     Subject: Optimize/OverOptimize
>   > 
>   > 
>   > 
>   >     Dave, I hope it's ok to contact you on this. steve karnish 
>   posted a
>   > presentation of yours on optimization that I found very 
>   interesting, though
>   > I'm afraid I don't get all of it. this is a topic I'm thinking 
>   about pretty
>   > much constantly these days, with quite a bit of accompanying 
>   frustration.
>   > IMVHO, most of the world gives way too much weight to 
optimizations 
>   that
>   > seem like curve fitting to me, but I haven't figured out how to 
>   move beyond
>   > that.
>   > 
>   > 
>   > 
>   >     a couple of questions, if I might:
>   > 
>   > 
>   > 
>   >     - can you explain the scatter plots on slides 3 and 4? what 
>   exactly is
>   > plotted on x and y? the punch line, which I'm too ignorant to 
see, 
>   is that
>   > the system fails with out of sample data. the one part I 
>   understand, I
>   > think, is that the correlation coefficient, presumably between 
in 
>   and out of
>   > sample results, is poor. is that right? how does the plot 
itself 
>   show this?
>   > 
>   > 
>   > 
>   >     They show the In-Sample gain as % of perfect trading on the 
x 
>   axis
>   > versus the out of sample gain on the y axis.  Each data point 
is a 
>   separate
>   > stock with a separate system.  In sample gains were 15% of 
perfect 
>   on
>   > average.  Out of sample were near zero on average.  Perfect 
trading 
>   wins all
>   > close to close changes.  There are 2 years in and out of sample.
>   > 
>   > 
>   > 
>   >     - slide 24 mentions "Trend Following on Commodities", 
as "100 
>   day
>   > lookback, trade 34% before breakout". I don't understand what 
this 
>   means.
>   > something about MA or EMA(100), maybe, but what's the 34% 
piece? 
>   how does it
>   > get around the parameter settings limitations that sink other 
>   systems? is
>   > this method, or something based on related principles, 
tradeable in 
>   stocks
>   > and/or mutual funds?
>   > 
>   > 
>   > 
>   >     Breakout buys a new high, sells a new low.  Near Breakout 
>   trades sooner.
>   > 34% before breakout buys in the top third of the 100 day high-
low 
>   range,
>   > sells in the bottom third.  Specifically, the 34% means 34% of 
the 
>   high-low
>   > range.
>   > 
>   > 
>   > 
>   >     - how would I compute the daily standard deviation of the 
>   S&P500, in
>   > AmiBroker for instance, in a way that gives the same .95%/day 
>   figure you
>   > mention? is that the average std dev of daily close price 
change 
>   over some
>   > specific period of time? I ask so I can generate comparable 
figures 
>   for
>   > other markets.
>   > 
>   > 
>   > 
>   >     Compute the standard deviation of all the close to close 
>   changes.
>   > 
>   > 
>   > 
>   > 
>   >     - the parameters I get optimizing today compensate for 
>   transient market
>   > behaviors that will eventually end, and eventually it will do 
very 
>   poorly.
>   > but if those behaviors persist, at least somewhat, for a little 
>   while, might
>   > the system to do better than average in the short term? if so, 
is 
>   constant
>   > re-optimization worth exploring, or even switching whole 
trading 
>   systems in
>   > a mechanical way based on recent performance?
>   > 
>   > 
>   > 
>   >     I find little tendency for trading systems to work in the 
>   future.  Try
>   > to identify a simple nonrandomness.  Try to find markets that 
>   simple systems
>   > work on.  Don't pick an impossible market like S&P 500 and try 
to 
>   fit a
>   > complex bunch of rules to it.
>   > 
>   > 
>   > 
>   >     Commodities have long term trends.  Stocks show short term 
2-10 
>   day
>   > reversals.
>   > 
>   > 
>   > 
>   >     thanks again for writing and sharing this. makes me wish I 
lived
>   > somewhere near the meetings you haunt...
>   > 
>   > 
>   > 
>   >     dave
>   > 
>   >       Dave is an Agilent, triple-degreed, engineer.  Two weeks 
ago, 
>   he
>   > presented this work to our Denver Trading Group's weekly 
meeting 
>   (actually,
>   > this group meets every Thursday and most Saturday's).  Once a 
>   month, I
>   > moderate a SIG on mechanical trading (and I haven't seen less 
than 
>   eighty
>   > people in the room since I've been attending).
>   > 
>   > 
>   > 
>   >       Although, I don't agree with certain aspects of his 
>   presentation and I
>   > somewhat object to his assigning my name to the "Karnish 
System" 
>   (it has
>   > become a bastardized off-shot of my work), I still believe that 
>   there is a
>   > lot of merit to aspects of his work.  The "Karnish System" has 
>   become the
>   > moniker for systems (along the front range of Colorado) that 
>   stochastically
>   > smoothes a momentum oscillator that initiates buy and sell 
signals 
>   using
>   > symmetrical triggers.
>   > 
>   > 
>   > 
>   >       I neither want to endorse, defend or criticize Dave's 
>   work...but,
>   > offer this for group members to stimulate thought.
>   > 
>   > 
>   > 
>   >       Take care,
>   > 
>   > 
>   > 
>   >       Steve
> 
> 
>         Yahoo! Groups Sponsor 
>        
>        
> 
>   Send BUG REPORTS to bugs@xxxx
>   Send SUGGESTIONS to suggest@xxxx
>   -----------------------------------------
>   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.


------------------------ Yahoo! Groups Sponsor ---------------------~-->
Rent DVDs Online - Over 14,500 titles.
No Late Fees & Free Shipping.
Try Netflix for FREE!
http://us.click.yahoo.com/JYdFFC/XP.FAA/ySSFAA/GHeqlB/TM
---------------------------------------------------------------------~->

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 http://docs.yahoo.com/info/terms/