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

[amibroker] Re: Monte Carlo analysis for trading systems



PureBytes Links

Trading Reference Links

> I know both David Aronson and Tim Masters.  I like and recommend 
>David's
> book, "Evidence-based Technical Analysis". 

I find that DA and TM's public work is the most challenging, and up 
to date material on system evaluation, going around i.e. for the 
general trading community (don't know what is happening in academia).

I am benchmarking my ideas against theirs.

I was giving Tims paper a careful re-reading yesterday and went to 
sleep (very quickly) with EBTA in my hand (no reflection on the book).

I don't think I will be going head to head with them any time soon, 
for obvious reasons, but I will be noting my concerns, about MCP as a 
tool for system evaluation, at the Zboard.

Naturally I will only do that in a naive way and won't be exhibiting 
the mathematical rigor, and testing, that TM does in his paper 
(others are welcome to do that, if they are interested, or refute my 
arguments in writing anywhere they like ... I will upload any quality 
posts mailed to me).

I am amazed at the low level of interest in the subject, by traders 
in general, and also that 'we' haven't moved along very quickly, 
since Pardo e.g. there's not a lot of quality books, on simulation 
for trading, available.

(Haven't read the Scherer and Martin book, recommended by Patrick 
yet). 

Please let me know of any other hardcore authors worth referencing.

> Tim's paper on
> Monte Carlo makes some assumptions that I think are inappropriate 
>for use
> when analyzing financial trading systems.

I am grateful that he made the effort and 'published' it.
It is the only definitive method I have found.

I also have some concerns about the method, albeit basic ones:

- it involves so many exceptions, to the extent that it is almost 
impractical for general trading applications (admittedly TM has 
provided a template that we can adjust to suit our own cirumstances)

- so far I am sceptical, about the possibility of mathematically 
detecting survivor bias in optimization runs etc.

Note that I don't claim 100% understanding of MCP, or Whites Reality 
Check, at this stage .... the fact that the MCP algorithm is written 
in a foreign language doesn't make it easy for me (I haven't looked 
at the C# version in our file section yet).

 > Two, I feel that Monte Carlo analysis is of limited
> value when the trading system is completely deterministic. 

I am not sure what you mean by 'deterministic' with regard to MCP.

Can you elaborate?


Re BiSim:

I feel it has some advantages over MCP and bootstrapping, at least as 
an eductational tool and possibly in some limited trading 
applications..... one of the advantages is that it is a convergence 
simulation (it approaches the mean outcome quickly and without 
massive effort) ... another is that it is very comfortable with 
correlation (in fact I think other 'modellers' make hard work out of 
handling it).

I hope to explore topics like that at the board.... obviously 
developing rigorous signinficance tests is going to be a challenge, 
if I get that far.

As I said, I am working live (I am not sure where it is going!), so 
if my future 'research' shoots down my own theories so be it.


Thanks for the feedback.


brian+z

--- In amibroker@xxxxxxxxxxxxxxx, Howard B <howardbandy@xxx> wrote:
>
> Hi Brian --
> 
> I know both David Aronson and Tim Masters.  I like and recommend 
David's
> book, "Evidence-based Technical Analysis".  Readers of both David's 
work and
> mine will find that David is even more conservative than I am 
regarding
> interpretation of in-sample versus out-of-sample results.  Tim's 
paper on
> Monte Carlo makes some assumptions that I think are inappropriate 
for use
> when analyzing financial trading systems.  Two points in 
particular.  One, I
> feel that neither bootstrapping nor jacknifing should be used when 
sampling
> financial time series.  Two, I feel that Monte Carlo analysis is of 
limited
> value when the trading system is completely deterministic.  I'll 
bring those
> up when I next talk with Tim.
> 
> Thanks,
> Howard
> 
> On Thu, Feb 12, 2009 at 4:54 PM, brian_z111 <brian_z111@xxx> wrote:
> 
> >   No rush.... I have been sitting on it for at least 2 years now.
> >
> > The full BinomialSimulation story won't be finished for months... 
I
> > will only post about once a month.
> >
> > First I am going to track back to the beginning, for the benefit 
of
> > non-mathematicians.
> >
> > Also, I will upload some stress test files, OR post images of the 
key
> > graphs from those files, so interested parties don't need to 
repeat
> > the massive simulations, for samples with bias and/or higher
> > dispersion, that I have already done.
> >
> > (Given your experience you would probably be best to sit back and
> > wait until I post the BS maths expression ... it will be very easy
> > for you to test and critique my theory at that stage ... you can
> > leave the hack work to me).
> >
> > Note that it is a work in progress i.e. I am working 'live', warts
> > and all, and I might not finish it, or leave it on the net, (I 
like
> > the Buddhist idea of 'pointing to the way' and demonstrating
> > impermanence).
> >
> > I probably won't 'advertise' here, in this forum, but BS posts 
will
> > go onto the Zboard blog page so they can get picked up by RSS.
> >
> > The other pages at the site, which are mainly just resources etc,
> > won't be disseminated via RSS.
> >
> > FYI I think BS is a significant method compared to bootstrapping 
and
> > MonteCarlo (considering their pros and cons) e.g. I disagree with
> > some of the assumptions of Timothy Masters, in his 2006 MCS 
article
> > at Aronsons 'EvidenceBasedTechnicalAnalysisSite' site... I also 
found
> > TM indecisive at some key points along the way.
> >
> > However, I am not going to follow the academic method of citing
> > others and criticising their work.
> >
> > I am very pleased you are looking at it.
> >
> > It has to stand up to the critique of informed mathematicians, 
like
> > yourself (more so than other new ideas because I am a naive
> > mathematician and an intuitive rather than a trained
> > objectivist/academic).
> >
> > Around 20 people downloaded the file .... some of them would be 
just
> > curious, or 'getting an education' (which are good things in
> > themselves) .... so at best there are only a few hard core 
analysts
> > considering my 'thesis'.
> >
> > Pity QT isn't still around ... he was a very nice guy and very 
good
> > on this stuff ;-)
> >
> > The Zboard site does allow for collaboration.
> >
> > If one or two self-managing people came along who wanted to add
> > something I could give them access ..... in that case the site 
would
> > stay online for the benefit of future googling traders who are in
> > search of trading truths.
> >
> > Cheers and thanks for your interest ... it's a compliment.
> >
> > brian.
> >
> > --- In amibroker@xxxxxxxxxxxxxxx <amibroker%40yahoogroups.com>, 
Howard B
> > <howardbandy@> wrote:
> > >
> > > Hi Brian --
> > >
> > > The zboard file worked fine.
> > >
> > > I have been snowed under with maintenance jobs the past week, so
> > it'll take
> > > me a couple of days to look at it.
> > >
> > > Thanks,
> > > Howard
> > >
> > > On Tue, Feb 10, 2009 at 1:06 AM, brian_z111 <brian_z111@> wrote:
> > >
> > > > Howard,
> > > >
> > > > I might move to MediaFire completely .. they are free
> > and 'permanent'
> > > > but the ads are terrible.
> > > >
> > > > With Rapidshare I will have to pay for some space to keep the
> > files
> > > > longer than 90 days but it is ad free.
> > > >
> > > > Haven't decided.
> > > >
> > > > Two files for you to try are at MF..... the PDF should give 
you a
> > > > quick test of the download.
> > > >
> > > > Refer to Mirror Site links:
> > > >
> > > > http://zboard.wordpress.com/downloads/
> > > >
> > > > Future:
> > > >
> > > > - may upload the stress test files
> > > > - I have a math method in mind to bypass the number crunching
> > > > - the math formula would make it pretty easy to do in AFL 
except
> > it
> > > > needs a trade array (workarounds possible with current AB 
version
> > I
> > > > guess)
> > > > - part 2 files explore sample error/variance (if they are 
going
> > > > somewhere I will post on that ... I recall I did find some
> > > > interesting relationships in error propogation but I haven't
> > looked
> > > > at it for a couple of years)
> > > >
> > > > Let me know if you can't download from mediafire
> > > >
> > > > OR if you can recommend a good filesharing site
> > > >
> > > > brian
> > > >
> > > >
> > > > --- In amibroker@xxxxxxxxxxxxxxx <amibroker%
40yahoogroups.com><amibroker%
> > 40yahoogroups.com>,
> > > > "brian_z111" <brian_z111@> wrote:
> > > > >
> > > > > The BS file is too big for Yahoo group files ... also it 
would
> > clog
> > > > > up limited space.
> > > > >
> > > > > I thought about AB third party but I have to 
download/maintain
> > > > third
> > > > > party software to FTP upload.... that annoys me somewhat (I 
am a
> > > > very
> > > > > independent type).
> > > > >
> > > > > The Zboard/WordPress arrangement is a trial ... if it goes
> > smoothly
> > > > I
> > > > > will keep it going for a while.
> > > > >
> > > > > I am happy with the WordPress (limited filetype/space)
> > arrangement,
> > > > > with a file host for sharing.
> > > > >
> > > > > So, now I will consider other filesharing hosts.
> > > > >
> > > > > Anyone you can download from?
> > > > >
> > > > > I can put one somewhere else for you.
> > > > >
> > > > >
> > > > > Don't worry I will make sure you get one, way or another.
> > > > >
> > > > > Better to get another host though because there will be at 
least
> > > > one
> > > > > more big file ..... if I keep going there might be plugins 
one
> > day
> > > > so
> > > > > I need a universal host.
> > > > >
> > > > >
> > > > > brian_z111
> > > > >
> > > > > Zboard.wordpress.com
> > > > >
> > > > >
> > > > >
> > > > > --- In amibroker@xxxxxxxxxxxxxxx <amibroker%
40yahoogroups.com><amibroker%
> > 40yahoogroups.com>,
> > Howard B
> > > > <howardbandy@> wrote:
> > > > > >
> > > > > > Hi Brian --
> > > > > >
> > > > > > I use a Hughes satellite connection to the Internet. It 
seems
> > > > that
> > > > > Hughes
> > > > > > appears to Rapidshare as a single user (which is always 
over
> > its
> > > > > limit), so
> > > > > > I am never able to download a Rapidshare file. If 
possible,
> > > > could
> > > > > you
> > > > > > upload the files to the Yahoo AmiBroker file section?
> > > > > >
> > > > > > Thanks,
> > > > > > Howard
> > > > > >
> > > > > >
> > > > > > On Sat, Feb 7, 2009 at 9:10 PM, brian_z111 <brian_z111@>
> > wrote:
> > > > > >
> > > > > > > I am using Rapidshare for file sharing.
> > > > > > >
> > > > > > > Free downloads are available but they are slower than 
paid
> > > > > download
> > > > > > > and limited to 1 download per time ... wait a while and 
you
> > can
> > > > > > > download again (still good value for my customers).
> > > > > > >
> > > > > > >
> > > > 
http://rapidshare.com/files/195190948/BinomialMaster_ABVersion.xls
> > > > > > >
> > > > > > > A short ReadMe, to help understand the file, is at:
> > > > > > >
> > > > > > > http://zboard.wordpress.com/
> > > > > > >
> > > > > > > I can answer a few questions about the details in the 
file
> > for a
> > > > > > > limited time (while my memory is fresh) .... post
> > questions, if
> > > > > any,
> > > > > > > via comments at the Zboard.
> > > > > > >
> > > > > > >
> > > > > > > --- In amibroker@xxxxxxxxxxxxxxx <amibroker%
40yahoogroups.com><amibroker%
> > 40yahoogroups.com><amibroker%
> > > > 40yahoogroups.com>,
> > > > > > > "brian_z111" <brian_z111@> wrote:
> > > > > > > >
> > > > > > > > File limits prevented me uploading the 
BinomialSimulation
> > file
> > > > > (s)
> > > > > > > to
> > > > > > > > this group ... 20MB per file. I will post links to at
> > least
> > > > one
> > > > > > > > example, at the following temporary site, sometime 
this
> > week:
> > > > > > > >
> > > > > > > > http://zboard.wordpress.com/
> > > > > > > >
> > > > > > > > I will post some basic notes afterall because the 
task of
> > > > > following
> > > > > > > > the Excel sheets would be beyond anyone without them.
> > > > > > > >
> > > > > > > > The site might live on for a while after that.
> > > > > > > >
> > > > > > > >
> > > > > > > > --- In amibroker@xxxxxxxxxxxxxxx <amibroker%
40yahoogroups.com><amibroker%
> > 40yahoogroups.com><amibroker%
> >
> > > > 40yahoogroups.com>,
> > > > > > > "brian_z111" <brian_z111@> wrote:
> > > > > > > > >
> > > > > > > > > I decided to post the Binomial Simulation files a 
few
> > days
> > > > > > > ago ...
> > > > > > > > I
> > > > > > > > > am not going to announce the upload so this post is 
the
> > > > > > > discussion
> > > > > > > > > link for them (one or more files will appear at some
> > stage).
> > > > > > > > >
> > > > > > > > > FTR They do predict the eq dist quite well, for 
biased
> > and
> > > > > none
> > > > > > > > > biased 'coins' but there is one thing about them 
that
> > does
> > > > > > > concern
> > > > > > > > > me ... I referenced the same synthetic trade series 
to
> > make
> > > > > the
> > > > > > > > > binomial distribution and to create the synthetic eq
> > > > > curves ...
> > > > > > > > that
> > > > > > > > > seems a bit incestuous in some ways.
> > > > > > > > >
> > > > > > > > > On the other hand they could be full of incorrect 
math
> > > > > > > assumptions
> > > > > > > > > cos I got the math off Wikipedia!
> > > > > > > > >
> > > > > > > > > Guru Brian ;-)
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > --- In amibroker@xxxxxxxxxxxxxxx<amibroker%
40yahoogroups.com><amibroker%
> > 40yahoogroups.com><amibroker%
> >
> > > > > 40yahoogroups.com>,
> > > > > > > "brian_z111" <brian_z111@>
> > > > > > > wrote:
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > > This is a valid model as long as stationarity
> > holds ...
> > > > I
> > > > > > > have
> > > > > > > > > > > simulated random trading 'systems' and 
predicted the
> > > > > outcome
> > > > > > > by
> > > > > > > > > > using
> > > > > > > > > > > binomial probability, that references a 
frequency
> > > > > > > distribution
> > > > > > > > of
> > > > > > > > > > the
> > > > > > > > > > > randomly generated trades, and it predicted the
> > actual
> > > > > equity
> > > > > > > > > > > distributions extremely well (a lognormal dist
> > appears
> > > > at
> > > > > > > very
> > > > > > > > > high
> > > > > > > > > > > N's).
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > More precisely, I have simulated trade series, 
using
> > the
> > > > > RNG in
> > > > > > > > > > Excel, for random walks (50/50 systems) and biased
> > > > systems,
> > > > > > > with
> > > > > > > > > > normally distributed trade series (I used
> > > > > CentralLimitThereom
> > > > > > > to
> > > > > > > > > > create NormDists from the uniform output of the
> > generator.
> > > > > > > > > >
> > > > > > > > > > I simulated equity curves, using the synthetic 
trades,
> > > > and
> > > > > at
> > > > > > > the
> > > > > > > > > > same time used BinomialProb to model the predicted
> > > > > distribution
> > > > > > > > of
> > > > > > > > > > the eq curves (I imagined I was tossing a coin 
with
> > > > variable
> > > > > > > > values
> > > > > > > > > > for heads and tails ... of course in trading we 
can
> > win
> > > > > lose or
> > > > > > > > > draw
> > > > > > > > > > whereas in my model we can only win or lose).
> > > > > > > > > >
> > > > > > > > > > You might like to see the files?
> > > > > > > > > >
> > > > > > > > > > I am bored with that topic.
> > > > > > > > > >
> > > > > > > > > > I am not a mathematician ... it might be a load 
of old
> > > > > rubbish
> > > > > > > > for
> > > > > > > > > > all I know.
> > > > > > > > > >
> > > > > > > > > > As our discussion shows .. we can't get any
> > statistical
> > > > > > > certainty
> > > > > > > > > > anywhere in trading ... only approximations and
> > > > > probabilties.
> > > > > > > > > >
> > > > > > > > > > It is just another approximation, like MCS and
> > involves
> > > > > massive
> > > > > > > > > > number crunching.
> > > > > > > > > >
> > > > > > > > > > I didn't finish it because I wanted a quick and 
dirty
> > > > > method.
> > > > > > > > > >
> > > > > > > > > > The files are rough as old bags.
> > > > > > > > > >
> > > > > > > > > > I didn't make notes so even I have a hard time
> > following
> > > > the
> > > > > > > > > > logic ... I had a look at them the other day I 
had to
> > > > start
> > > > > > > > tracing
> > > > > > > > > > the formulas in the cells to see how I had done 
it.
> > > > > > > > > >
> > > > > > > > > > I'll post some of them in the file section one day
> > (Howard
> > > > > > > > collects
> > > > > > > > > > trading things).
> > > > > > > > > >
> > > > > > > > > > I won't scrub them up though ... take them or 
leave
> > > > them ...
> > > > > > > > sorry
> > > > > > > > > no
> > > > > > > > > > questions or explanations (anyway Howard and other
> > maths
> > > > > people
> > > > > > > > > know
> > > > > > > > > > how to do that stuff).
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > --- In amibroker@xxxxxxxxxxxxxxx<amibroker%
40yahoogroups.com><amibroker%
> > 40yahoogroups.com><amibroker%
> >
> > > > > 40yahoogroups.com>,
> > > > > > > "brian_z111" <brian_z111@>
> > > > > > > > wrote:
> > > > > > > > > > >
> > > > > > > > > > > Gidday Mate,
> > > > > > > > > > >
> > > > > > > > > > > I wasn't planning on posting again today as I am
> > going
> > > > > away
> > > > > > > for
> > > > > > > > a
> > > > > > > > > > few
> > > > > > > > > > > days ..... a good question though so I couldn't
> > resist.
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > I did notice Fred's comment on the priority he
> > places on
> > > > > > > > > > sensitivity
> > > > > > > > > > > analysis.
> > > > > > > > > > >
> > > > > > > > > > > He has made the comment before and I came to 
that
> > view
> > > > > > > > > > independently
> > > > > > > > > > > a way back anyway (Howard's random noise test is
> > another
> > > > > > > > > > interesting
> > > > > > > > > > > idea for single sample analysis).
> > > > > > > > > > >
> > > > > > > > > > > I also recall that he doesn't believe 
scrambling the
> > > > > order of
> > > > > > > > the
> > > > > > > > > > > trades provides any meaningful feedback.
> > > > > > > > > > >
> > > > > > > > > > > That isn't a reason for me not to reach my own
> > > > > conclusions.
> > > > > > > > > > >
> > > > > > > > > > > Fred has also talked about small N retesting 
(walk
> > > > > forward),
> > > > > > > > and
> > > > > > > > > > > adjusting his system rules, on a short term 
basis,
> > so
> > > > > while I
> > > > > > > > am
> > > > > > > > > > not
> > > > > > > > > > > keen on the idea I am keeping an open mind on 
the
> > > > subject.
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > > This is the second time in the >past few
> > > > > > > > > > > > days that you seem to have equated
> > trading/backtesting
> > > > > > > system
> > > > > > > > > > > >outcomes
> > > > > > > > > > > > to a random series of coin flip outcomes 
(random
> > > > binary
> > > > > > > > > > occurances).
> > > > > > > > > > > >
> > > > > > > > > > > > Serious question... what is your point? What 
is
> > the
> > > > > > > > relevence
> > > > > > > > > os
> > > > > > > > > > > >the
> > > > > > > > > > > > "Coin Flip" metaphor where trading systems is
> > > > concerned?
> > > > > > > > > > >
> > > > > > > > > > > Well, developers are selling software 
specifically
> > > > > designed
> > > > > > > for
> > > > > > > > > > > performing MSC for trading analysis and at 
least one
> > > > guy
> > > > > has
> > > > > > > > > > written
> > > > > > > > > > > a book on the subject.
> > > > > > > > > > >
> > > > > > > > > > > In both software packages, that I have some
> > familiarity
> > > > > with,
> > > > > > > > > their
> > > > > > > > > > > model assumes stationarity, and independency 
i.e.
> > their
> > > > > model
> > > > > > > > > > treats
> > > > > > > > > > > the data as if it is the outcome of a coin toss 
with
> > > > > variable
> > > > > > > > > > values
> > > > > > > > > > > on the +- side of the coin.
> > > > > > > > > > >
> > > > > > > > > > > This is a valid model as long as stationarity
> > holds ...
> > > > I
> > > > > > > have
> > > > > > > > > > > simulated random trading 'systems' and 
predicted the
> > > > > outcome
> > > > > > > by
> > > > > > > > > > using
> > > > > > > > > > > binomial probability, that references a 
frequency
> > > > > > > distribution
> > > > > > > > of
> > > > > > > > > > the
> > > > > > > > > > > randomly generated trades, and it predicted the
> > actual
> > > > > equity
> > > > > > > > > > > distributions extremely well (a lognormal dist
> > appears
> > > > at
> > > > > > > very
> > > > > > > > > high
> > > > > > > > > > > N's).
> > > > > > > > > > >
> > > > > > > > > > > The value, to me in that model, is that it is a
> > > > training
> > > > > tool
> > > > > > > > > that
> > > > > > > > > > > conditioned me to accept variance as 'normal' 
and
> > if the
> > > > > > > market
> > > > > > > > > is
> > > > > > > > > > > stationary then it would have direct relevance 
to
> > > > > > > trading.....
> > > > > > > > > the
> > > > > > > > > > > worst case outcome would be that I could incur
> > losses,
> > > > > with a
> > > > > > > > > > > probability as indicated by the Cumulative
> > Distrubution
> > > > > > > > Function
> > > > > > > > > > for
> > > > > > > > > > > the possible equity outcomes (simulation is one 
way
> > for
> > > > > non -
> > > > > > > > > > > mathematicians to calc this and view it in a 
chart).
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > Ask yourself ....
> > > > > > > > > > >
> > > > > > > > > > > afer you have conducted a successful OOS, and
> > collated
> > > > the
> > > > > > > > trade
> > > > > > > > > > > sample, when you start to trade it do you 
expect:
> > > > > > > > > > >
> > > > > > > > > > > - all trades to be the same, or similar, and 
occur
> > with
> > > > > the
> > > > > > > > same
> > > > > > > > > > > frequency (TradeSim),
> > > > > > > > > > > - all trades to be the same, or similar, and 
have
> > > > > variations
> > > > > > > in
> > > > > > > > > the
> > > > > > > > > > > frequency (MSA),
> > > > > > > > > > > - something else?
> > > > > > > > > > >
> > > > > > > > > > > Trading, however, is not a coin toss.
> > > > > > > > > > >
> > > > > > > > > > > It is more like a sample generator that produces
> > trades
> > > > > as a
> > > > > > > > > result
> > > > > > > > > > > of presenting dynamic data to the system 
(filter).
> > > > > > > > > > >
> > > > > > > > > > > To what extent could a 'real life' trading 
system
> > > > emulate
> > > > > a
> > > > > > > > coin
> > > > > > > > > > > toss, with variable values ... how could that 
come
> > > > about?
> > > > > > > > > > >
> > > > > > > > > > > (interesting that the very functional optF 
formula
> > came
> > > > > about
> > > > > > > > as
> > > > > > > > > > the
> > > > > > > > > > > variable value coin toss staking formula).
> > > > > > > > > > >
> > > > > > > > > > > Is it possible or not?
> > > > > > > > > > >
> > > > > > > > > > > A lot of people seem to think it is, judging by
> > their
> > > > > books
> > > > > > > and
> > > > > > > > > > > software.
> > > > > > > > > > >
> > > > > > > > > > > Presumably, when the underlying data changes, 
the
> > sample
> > > > > > > > profile
> > > > > > > > > > > (mean, StDev etc) can change and we end up with 
a
> > > > better
> > > > > or
> > > > > > > > worse
> > > > > > > > > > > outcome than anticipated by the OOS.
> > > > > > > > > > >
> > > > > > > > > > > So, does the non-stationary behaviour of the 
markets
> > > > > > > invalidate
> > > > > > > > > the
> > > > > > > > > > > coin toss model?
> > > > > > > > > > >
> > > > > > > > > > > That is the ineresting question, and I don't 
know
> > the
> > > > > answer
> > > > > > > to
> > > > > > > > > it,
> > > > > > > > > > > or even if there is a definite answer.
> > > > > > > > > > >
> > > > > > > > > > > I was hopeful that people would pick up on that 
key
> > > > point
> > > > > and
> > > > > > > > > shed
> > > > > > > > > > > some light on the subject.
> > > > > > > > > > >
> > > > > > > > > > > I know, from my long hours of simulating random
> > data,
> > > > what
> > > > > > > > random
> > > > > > > > > > > behaviour looks like when I see it.
> > > > > > > > > > >
> > > > > > > > > > > Clearly the markets have a certain amount of 
random
> > > > > behaviour.
> > > > > > > > > > >
> > > > > > > > > > > Howard commented somewhere, or another, that 
there
> > is a
> > > > > > > certain
> > > > > > > > > > > amount of randomness in the market (I can't 
recall
> > the
> > > > > method
> > > > > > > > he
> > > > > > > > > > used
> > > > > > > > > > > to measure it).
> > > > > > > > > > >
> > > > > > > > > > > It is quite easy to observe if data has any 
random
> > > > > qualities,
> > > > > > > > > > > especially if we measure the core attributes 
(50/50
> > > > heads
> > > > > and
> > > > > > > > > tails
> > > > > > > > > > > and its persistence into 2,3,4 heads in a row 
etc).
> > > > > > > > > > >
> > > > > > > > > > > Once again I ask you to consider:
> > > > > > > > > > >
> > > > > > > > > > > if I measure the S&P500 index, on close, and it
> > goes up
> > > > > > > approx
> > > > > > > > 50
> > > > > > > > > > and
> > > > > > > > > > > down approx 50 (+- variance that is typical of a
> > random
> > > > > > > > binomial
> > > > > > > > > > > event) and the subsequent second head or tail 
follow
> > > > with
> > > > > 0.5
> > > > > > > > > prob
> > > > > > > > > > > etc I am justified in considering it top be a 
pseudo
> > > > > random
> > > > > > > > > > binomail
> > > > > > > > > > > event?
> > > > > > > > > > >
> > > > > > > > > > > I have done quick and dirty measurements, and
> > accurate
> > > > > > > > > > measurements,
> > > > > > > > > > > on dependency (or on its inverse, which is
> > > > independency)
> > > > > and
> > > > > > > > find
> > > > > > > > > > > that there is a good deal of independency in the
> > > > markets
> > > > > (I
> > > > > > > > > posted
> > > > > > > > > > > some q&d code to measure that last week).
> > > > > > > > > > >
> > > > > > > > > > > I have speculated before, on the point, that the
> > > > rational
> > > > > > > > market
> > > > > > > > > is
> > > > > > > > > > > the market that follows fundamental value, which
> > tends
> > > > to
> > > > > be
> > > > > > > >=
> > > > > > > > > the
> > > > > > > > > > > yearly (macro) timeframe, and, everything else 
is
> > the
> > > > > > > > irrational
> > > > > > > > > > > market.
> > > > > > > > > > >
> > > > > > > > > > > Consider an intraday market ... what is rational
> > about
> > > > the
> > > > > > > > price
> > > > > > > > > > > movement during any given part of the day?
> > > > > > > > > > >
> > > > > > > > > > > - Draw a trend line on the chart .. we will 
assume
> > that
> > > > we
> > > > > > > know
> > > > > > > > > > what
> > > > > > > > > > > a trend is for this exercise, although that is a
> > > > debatable
> > > > > > > > point.
> > > > > > > > > > >
> > > > > > > > > > > - The trend, a straight line, is rational (it is
> > > > perfectly
> > > > > > > > > > following
> > > > > > > > > > > fundamental value).... it is 2007 and it is 
up ;-)
> > > > > > > > > > >
> > > > > > > > > > > - All of the ups and downs that occur around it 
are
> > > > > > > irrational
> > > > > > > > > > > (bucking the trend).
> > > > > > > > > > >
> > > > > > > > > > > - The trend line goes under the pivot lows.
> > > > > > > > > > >
> > > > > > > > > > > - Your system buys at the pivot lows and sells 
at =
> > = 2
> > > > > StDev
> > > > > > > > > above
> > > > > > > > > > > the trend line.
> > > > > > > > > > >
> > > > > > > > > > > - Place a stop under the trend line at - 1 
stDev.
> > > > > > > > > > >
> > > > > > > > > > > - Assume no commission and no slippage.
> > > > > > > > > > >
> > > > > > > > > > > - Your payoff ratio is 2/1
> > > > > > > > > > >
> > > > > > > > > > > - assume there is no variance in volatility so 
the
> > PR
> > > > is a
> > > > > > > > > constant
> > > > > > > > > > > value
> > > > > > > > > > >
> > > > > > > > > > > - the win/loss ratio is determined by the random
> > > > > meandering
> > > > > > > of
> > > > > > > > > the
> > > > > > > > > > > irrational price movements up and down.
> > > > > > > > > > >
> > > > > > > > > > > Note they are irrational because people are 
buying
> > and
> > > > > > > selling
> > > > > > > > at
> > > > > > > > > > the
> > > > > > > > > > > wrong time and for the wrong reasons - if they 
were
> > > > > rational
> > > > > > > > they
> > > > > > > > > > > would only be buying selling as fundamental 
values
> > > > change.
> > > > > > > > > > >
> > > > > > > > > > > - the trade series produced would look exactly 
that
> > that
> > > > > > > > produced
> > > > > > > > > > by
> > > > > > > > > > > a coin tossed with +2, -1 value on it.
> > > > > > > > > > >
> > > > > > > > > > > Now, you have tested this system, OOS, and it 
is a
> > > > winner.
> > > > > > > > > > >
> > > > > > > > > > > What chance for stationarity when you trade 
live?
> > > > > > > > > > >
> > > > > > > > > > > If the trend continues there is a very good 
chance
> > that
> > > > > the
> > > > > > > > > random
> > > > > > > > > > > emualator (system meeting dynamic data) will
> > continue to
> > > > > > > > perform
> > > > > > > > > > like
> > > > > > > > > > > a biased coin +- variance i.e. the payoff ratio
> > can't
> > > > > change
> > > > > > > > but
> > > > > > > > > > the
> > > > > > > > > > > W/L will (it always does when I toss a coin).
> > > > > > > > > > >
> > > > > > > > > > > If the trend changes your winning model will be 
more
> > > > > likely
> > > > > > > to
> > > > > > > > > bust.
> > > > > > > > > > >
> > > > > > > > > > > That could be the reason Fred, and others, like 
to
> > > > > > > continually
> > > > > > > > > > retest.
> > > > > > > > > > >
> > > > > > > > > > > I have another approach to getting around this
> > problem
> > > > > (this
> > > > > > > is
> > > > > > > > > > > actually the real point of my posts) ...
> > > > > > > > > > >
> > > > > > > > > > > ..... to accomodate non-stationarity either 
adjust
> > > > > quickly OR
> > > > > > > > use
> > > > > > > > > a
> > > > > > > > > > > dimensionless model e.g. don't believe in 
trends and
> > > > then
> > > > > you
> > > > > > > > > can't
> > > > > > > > > > > be on the wrong side of them.
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > However, that is only speculation.
> > > > > > > > > > >
> > > > > > > > > > > What do you think?
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > Again ... what is the relevance of coin tosses 
to
> > > > trading
> > > > > IMO:
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > - wonderful training tool
> > > > > > > > > > > - a good OOS can not predict exactly what the
> > outcome
> > > > of
> > > > > live
> > > > > > > > > > trading
> > > > > > > > > > > will be (subject to nonstationarity) and 
neither can
> > > > > > > simulation
> > > > > > > > > > (coin
> > > > > > > > > > > tossing) but it gives a good approximation of 
the
> > > > > > > possibilities
> > > > > > > > > > (also
> > > > > > > > > > > subject to non-stationarity).
> > > > > > > > > > >
> > > > > > > > > > > As a quid pro quo .....
> > > > > > > > > > >
> > > > > > > > > > > ..... if you, or anyone else, can give me any
> > > > explanation
> > > > > > > > and/or
> > > > > > > > > > > proof that the coin toss metaphor has no 
relevance
> > to
> > > > > trading
> > > > > > > I
> > > > > > > > > > would
> > > > > > > > > > > be delighted.
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > Anyway, I think Patrick already answered the
> > question,
> > > > or
> > > > > > > told
> > > > > > > > us
> > > > > > > > > > > where to find it.
> > > > > > > > > > >
> > > > > > > > > > > Good luck with your trading.
> > > > > > > > > > >
> > > > > > > > > > > brian_zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > --- In amibroker@xxxxxxxxxxxxxxx<amibroker%
40yahoogroups.com>
> > <amibroker%
> > 40yahoogroups.com><amibroker%
> >
> > > > > 40yahoogroups.com>,
> > > > > > > "Phsst" <phsst@> wrote:
> > > > > > > > > > > >
> > > > > > > > > > > > Hello Brian,
> > > > > > > > > > > >
> > > > > > > > > > > > Thanks for the mention in your New Years 
post. I
> > felt
> > > > > > > > humbled
> > > > > > > > > to
> > > > > > > > > > > be in
> > > > > > > > > > > > the same honerable mention list as Fred (He 
is a
> > very
> > > > > smart
> > > > > > > > > Dude
> > > > > > > > > > (no
> > > > > > > > > > > > kidding!)) It took me a while (some years 
back) to
> > > > > figure
> > > > > > > out
> > > > > > > > > > what a
> > > > > > > > > > > > smart guy Fred really is. I've since learned 
that
> > > > when
> > > > > Fred
> > > > > > > > > > speaks,
> > > > > > > > > > > it
> > > > > > > > > > > > pays to think and be silent for a good long 
while
> > > > before
> > > > > > > > > drawing
> > > > > > > > > > any
> > > > > > > > > > > > conclusions.
> > > > > > > > > > > >
> > > > > > > > > > > > To your "crystal clear" point... This is the
> > second
> > > > > time in
> > > > > > > > the
> > > > > > > > > > > past few
> > > > > > > > > > > > days that you seem to have equated
> > trading/backtesting
> > > > > > > system
> > > > > > > > > > > outcomes
> > > > > > > > > > > > to a random series of coin flip outcomes 
(random
> > > > binary
> > > > > > > > > > occurances).
> > > > > > > > > > > >
> > > > > > > > > > > > Serious question... what is your point? What 
is
> > the
> > > > > > > > relevence
> > > > > > > > > os
> > > > > > > > > > > the
> > > > > > > > > > > > "Coin Flip" metaphor where trading systems is
> > > > concerned?
> > > > > > > > What
> > > > > > > > > am
> > > > > > > > > > I
> > > > > > > > > > > > missing?
> > > > > > > > > > > >
> > > > > > > > > > > > Your Bud... Phsst
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > This is the second time
> > > > > > > > > > > > --- In amibroker@xxxxxxxxxxxxxxx<amibroker%
40yahoogroups.com>
> > <amibroker%
> > 40yahoogroups.com><amibroker%
> >
> > > > > 40yahoogroups.com>,
> > > > > > > "brian_z111"
> > > > > > > <brian_z111@>
> > > > > > > > > > wrote:
> > > > > > > > > > > > >
> > > > > > > > > > > > > To be chrystal clear about my hypothesis:
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > > > We are trying to design a system that 
produces
> > the
> > > > > same
> > > > > > > set
> > > > > > > > of
> > > > > > > > > > > > > trades, in the future, as it has in the 
past i.e
> > > > > trades
> > > > > > > and
> > > > > > > > > not
> > > > > > > > > > > > > combinations of trades.
> > > > > > > > > > > > >
> > > > > > > > > > > > > If a solid gold coin, minted by the US 
treasury,
> > > > with
> > > > > a
> > > > > > > > head
> > > > > > > > > > and a
> > > > > > > > > > > > > tail clearly stamped on each side, and only 
two
> > > > > values +1
> > > > > > > > or -
> > > > > > > > > 1
> > > > > > > > > > > can't
> > > > > > > > > > > > > reproduce two equity curves that look the 
same,
> > > > after
> > > > > N
> > > > > > > > > tosses,
> > > > > > > > > > > how
> > > > > > > > > > > > > can we expect a trading system to do that 
when
> > it
> > > > has
> > > > > a
> > > > > > > > range
> > > > > > > > > of
> > > > > > > > > > > > > possible values?
> > > > > > > > > > > > >
> > > > > > > > > > > > > AND it doesn't get any better as N 
increases.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Put your time and effort into maximising the
> > > > STABILITY
> > > > > > > > > > > > > (predictability, boundness) of the trade
> > set 'with
> > > > an
> > > > > > > edge'
> > > > > > > > > > THEN
> > > > > > > > > > > use
> > > > > > > > > > > > > MM to optimise the equity outcome the system
> > > > produces
> > > > > > > > > (optimise
> > > > > > > > > > ==
> > > > > > > > > > > > > your definition e.g. max return, min risk or
> > > > > whatever).
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > > > --- In amibroker@xxxxxxxxxxxxxxx<amibroker%
40yahoogroups.com>
> > <amibroker%
> > 40yahoogroups.com><amibroker%
> >
> > > > > 40yahoogroups.com>,
> > > > > > > "brian_z111"
> > > > > > > brian_z111@
> > > > > > > > > > wrote:
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Howard,
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Thanks for your post.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > A very well written article.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Some contrary comment (first referencing 
some
> > of
> > > > > your
> > > > > > > > > points
> > > > > > > > > > and
> > > > > > > > > > > > > > then, later, some comments of my own):
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > > By trying many
> > > > > > > > > > > > > > > combinations of logic and parameter 
values,
> > we
> > > > > will
> > > > > > > > > > eventually
> > > > > > > > > > > > > find
> > > > > > > > > > > > > > >a system that is profitable for the date
> > range
> > > > > > > analyzed.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > You are assuming that all successful long 
term
> > > > > traders
> > > > > > > > > > arrived
> > > > > > > > > > > at
> > > > > > > > > > > > > > their system(s) by using this approach ...
> > perhaps
> > > > > > > there
> > > > > > > > > are
> > > > > > > > > > > > > systems
> > > > > > > > > > > > > > out there that have no optimiseable 
parameters
> > > > and
> > > > > only
> > > > > > > > one
> > > > > > > > > > > > > > underlying logic.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > If so they are likely be based on primal
> > market
> > > > > > > behaviour
> > > > > > > > > and
> > > > > > > > > > > > > > therefore persistent across markets and 
time
> > i.e
> > > > > they
> > > > > > > > would
> > > > > > > > > > > have to
> > > > > > > > > > > > > > be systems based on market characteristics
> > that
> > > > are
> > > > > > > > > relatively
> > > > > > > > > > > > > > stationary.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > > testing the
> > > > > > > > > > > > > > > profitability of a trading system that 
was
> > > > > developed
> > > > > > > > > using
> > > > > > > > > > > recent
> > > > > > > > > > > > > > >data
> > > > > > > > > > > > > > > on older data is guaranteed to over-
> > estimate the
> > > > > > > > > > > profitability of
> > > > > > > > > > > > > > the
> > > > > > > > > > > > > > > trading system.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > You know that in science 
(philosophy/logic) it
> > > > only
> > > > > > > takes
> > > > > > > > > one
> > > > > > > > > > > > > > refutation to dethrone the current ruling
> > > > > hypothesis ...
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > if a long system, developed on the last 12
> > months
> > > > of
> > > > > > > data
> > > > > > > > > > (when
> > > > > > > > > > > the
> > > > > > > > > > > > > > market was experiencing a bear riot) is 
then
> > > > tested
> > > > > OOS
> > > > > > > > on
> > > > > > > > > the
> > > > > > > > > > > > > prior
> > > > > > > > > > > > > > years data it will outperform the in 
sample
> > tests
> > > > > (OOS
> > > > > > > > > would
> > > > > > > > > > be
> > > > > > > > > > > > > > conducted on bull market data).
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > > There is very little reason to expect 
that
> > > > future
> > > > > > > > > behavior
> > > > > > > > > > and
> > > > > > > > > > > > > > > profitability of well known trading 
systems
> > > > will
> > > > > be
> > > > > > > the
> > > > > > > > > > same
> > > > > > > > > > > as
> > > > > > > > > > > > > past
> > > > > > > > > > > > > > > behavior.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Do we have any empirical evidence of this?
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > First we would have to have an agreed
> > definition
> > > > > > > of 'well
> > > > > > > > > > > known',
> > > > > > > > > > > > > > make a list of the systems, and then 
perform
> > > > massive
> > > > > > > > > testing.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > To scrupulously prevent any bias creeping
> > testing
> > > > > would
> > > > > > > > > have
> > > > > > > > > > to
> > > > > > > > > > > be
> > > > > > > > > > > > > > conducted live, and not on historical 
data.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > We only know that they were 
successful 'in the
> > > > > past' by
> > > > > > > > IS
> > > > > > > > > > > testing,
> > > > > > > > > > > > > > or by claim.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Do we have any, or many, certified 
performance
> > > > > records
> > > > > > > > > > provided
> > > > > > > > > > > by
> > > > > > > > > > > > > > traders who claim to have had success with
> > > > > those 'well
> > > > > > > > > known'
> > > > > > > > > > > > > systems.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Statistics gathered from in-sample 
results
> > have
> > > > > > > > > > > > > > > no relationship to statistics that will 
be
> > > > > gathered
> > > > > > > > from
> > > > > > > > > > > trading.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Not, so.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > They have every bearing on the stats 
gathered
> > in
> > > > > > > trading
> > > > > > > > > > because
> > > > > > > > > > > > > only
> > > > > > > > > > > > > > systems with good IS performance make it 
to
> > the
> > > > OS,
> > > > > or
> > > > > > > > live
> > > > > > > > > > > > > trading,
> > > > > > > > > > > > > > phase.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > OOS testing is only proceeded with 
because the
> > > > > analyst
> > > > > > > > has
> > > > > > > > > > every
> > > > > > > > > > > > > > expectation, or hope, that the good IS 
stats
> > will
> > > > be
> > > > > > > > > > reproduced
> > > > > > > > > > > OOS.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > In fact it is the relative performance 
between
> > > > the
> > > > > IS
> > > > > > > and
> > > > > > > > > OOS
> > > > > > > > > > > stats
> > > > > > > > > > > > > > the encourages us to proceed or abort.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Re trading the edge erodes the edge:
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > It is an assumption that all players are
> > trading
> > > > > > > > > systems ...
> > > > > > > > > > > many
> > > > > > > > > > > > > are
> > > > > > > > > > > > > > not, in fact the vast majority are not....
> > those
> > > > who
> > > > > > > > aren't
> > > > > > > > > > > control
> > > > > > > > > > > > > > vastly greater sums of money than those 
who
> > do.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > It is an assumption that all wins erode 
the
> > > > > system ...
> > > > > > > > they
> > > > > > > > > > > could
> > > > > > > > > > > > > be
> > > > > > > > > > > > > > just lucky wins that the trader can't 
exploit
> > long
> > > > > > > term,
> > > > > > > > or
> > > > > > > > > > > > > > successful wins that the trader doesn't
> > sustain
> > > > e.g
> > > > > > > they
> > > > > > > > > > might
> > > > > > > > > > > not
> > > > > > > > > > > > > > have the capital, use the correct staking 
or
> > > > > maintain
> > > > > > > > self-
> > > > > > > > > > > > > discipline
> > > > > > > > > > > > > > in the future.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Only a very small percentage of traders 
are
> > > > > successful,
> > > > > > > > and
> > > > > > > > > > > hence
> > > > > > > > > > > > > > trading a successful system ... every one 
else
> > > > who
> > > > > is
> > > > > > > > > trading
> > > > > > > > > > is
> > > > > > > > > > > > > just
> > > > > > > > > > > > > > making noise.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > There are millions of system permutations,
> > > > > instruments,
> > > > > > > > > > markets,
> > > > > > > > > > > > > > staking systems etc ..... how many 
successful
> > > > > traders
> > > > > > > > would
> > > > > > > > > > it
> > > > > > > > > > > take
> > > > > > > > > > > > > > to exahaust all of the successful
> > permutations?
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > > The follow-on point, which relates to 
Monte
> > > > Carlo
> > > > > > > > > analysis,
> > > > > > > > > > is
> > > > > > > > > > > > > that
> > > > > > > > > > > > > > > rearranging the in-sample trades gives 
no
> > > > insight
> > > > > > > into
> > > > > > > > > the
> > > > > > > > > > > future
> > > > > > > > > > > > > > > characteristics of the system. Yes, you 
can
> > see
> > > > > the
> > > > > > > > > effect
> > > > > > > > > > of
> > > > > > > > > > > > > taking
> > > > > > > > > > > > > > > the trades in different orders. But why
> > bother?
> > > > > They
> > > > > > > > are
> > > > > > > > > > still
> > > > > > > > > > > > > > > in-sample results and still have no 
value.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > If you are engineering an F1 racing car 
there
> > is
> > > > > only
> > > > > > > > track
> > > > > > > > > > > > > > testing/simulation (99.9 of the time) and
> > racing
> > > > > > > > > performance
> > > > > > > > > > > (1% of
> > > > > > > > > > > > > > the time).
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > The more information you gather off the 
track
> > the
> > > > > more
> > > > > > > > > likely
> > > > > > > > > > > you
> > > > > > > > > > > > > are
> > > > > > > > > > > > > > to perform on the track OR know what to 
adjust
> > > > and
> > > > > when
> > > > > > > > to
> > > > > > > > > > > adjust
> > > > > > > > > > > > > it
> > > > > > > > > > > > > > if performance doesn't meet expectations.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Do you know of any F1 teams that don't
> > > > > test/simulate?
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Do you know of any F1 teams that only
> > > > test/simulate
> > > > > > > one,
> > > > > > > > or
> > > > > > > > > > > > > limited,
> > > > > > > > > > > > > > metrics?
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > What is testing if not 'massive 
examination of
> > > > what-
> > > > > if
> > > > > > > > > > > scenarios'?
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Re MonteCarlo and stationarity
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > I haven't studied the subject in depth.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Mainly it is has been used outside of 
trading
> > and
> > > > in
> > > > > > > > > > different
> > > > > > > > > > > ways
> > > > > > > > > > > > > > to the ways that traders use it .... 
possibly
> > it
> > > > > would
> > > > > > > > be
> > > > > > > > > > best
> > > > > > > > > > > to
> > > > > > > > > > > > > > limit trading discussion to 'trading
> > simulation'
> > > > and
> > > > > > > drop
> > > > > > > > > the
> > > > > > > > > > MC
> > > > > > > > > > > > > part
> > > > > > > > > > > > > > of the name.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > I have only found one book devoted to the
> > subject
> > > > > and I
> > > > > > > > > regret
> > > > > > > > > > > > > buying
> > > > > > > > > > > > > > it .... 'MCS and System Trading' by Volker
> > > > Butzlaff.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > I have also test driven TradeSim and MSA.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Referencing their trading apps.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > TS arranges the trades, as a time series, 
and
> > > > > randomly
> > > > > > > > > walks
> > > > > > > > > > > > > through
> > > > > > > > > > > > > > all permutations to simulate 'live
> > trading'.....
> > > > it
> > > > > is
> > > > > > > an
> > > > > > > > > MM
> > > > > > > > > > > test,
> > > > > > > > > > > > > of
> > > > > > > > > > > > > > some kind, because equity is allocated 
prior
> > to
> > > > the
> > > > > > > walk
> > > > > > > > > > > through.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > AB's backtester, in default mode, does 
this
> > once.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > I assume other methods could be used ... 
as
> > per my
> > > > > > > > pervious
> > > > > > > > > > XYZ
> > > > > > > > > > > > > > example:
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > - abcXdefghi with simultaneous trades on 
day
> > 4,
> > > > > > > > > > > > > > - we can only achieve a finite set of
> > > > permutations,
> > > > > > > > > > > > > > - the outcome of massive sampling will 
tend to
> > > > the
> > > > > mean
> > > > > > > +-
> > > > > > > >
> > > > > > > > > > > variance,
> > > > > > > > > > > > > > - we can simulate the eq outcomes using 
random
> > > > > sampling
> > > > > > > > of
> > > > > > > > > > > uniform
> > > > > > > > > > > > > > size, ave the result per random series and
> > then
> > > > freq
> > > > > > > dist
> > > > > > > > > the
> > > > > > > > > > > means
> > > > > > > > > > > > > > (Central Limit Theoreom predicts a pseudo 
norm
> > > > > dist).
> > > > > > > > > > > > > > > 30 selections per series * ? series will
> > > > achieve
> > > > > an
> > > > > > > > > approx
> > > > > > > > > > of
> > > > > > > > > > > > > > possible eq outcomes (I'm not sure if
> > > > distrubtions
> > > > > obey
> > > > > > > > the
> > > > > > > > > > > laws of
> > > > > > > > > > > > > > sample error ... I don't think they do).
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > TradeSims real life simulation assumes
> > > > stationarity
> > > > > > > (the
> > > > > > > > > > balls
> > > > > > > > > > > in
> > > > > > > > > > > > > the
> > > > > > > > > > > > > > bin, and their values will remain constant
> > into
> > > > the
> > > > > > > > future).
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > It also assumes that they will be selected
> > from
> > > > the
> > > > > bin
> > > > > > > > in
> > > > > > > > > > the
> > > > > > > > > > > same
> > > > > > > > > > > > > > order, or frequency to be absolutely 
correct
> > (the
> > > > > order
> > > > > > > > > > doesn't
> > > > > > > > > > > > > > change anything only the frequency).... 
to be
> > > > > precise
> > > > > > > > about
> > > > > > > > > > it,
> > > > > > > > > > > > > their
> > > > > > > > > > > > > > model assumes that if you have picked the
> > worst
> > > > > > > > historical
> > > > > > > > > > loss
> > > > > > > > > > > out
> > > > > > > > > > > > > > of the bin 2/1000 trades that you will not
> > only
> > > > > > > > experience
> > > > > > > > > > the
> > > > > > > > > > > same
> > > > > > > > > > > > > %
> > > > > > > > > > > > > > as the worst loss in the future but that 
it
> > will
> > > > > also
> > > > > > > > only
> > > > > > > > > > occur
> > > > > > > > > > > > > > 2/1000 times.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > MSA puts all of the balls in the bin and
> > selects
> > > > > them
> > > > > > > in
> > > > > > > > a
> > > > > > > > > > way
> > > > > > > > > > > that
> > > > > > > > > > > > > > allows new combinations (frequencies) 
until
> > all
> > > > > possible
> > > > > > > > > > > > > frequencies
> > > > > > > > > > > > > > are exhausted i.e. they assume 
stationarity
> > only
> > > > in
> > > > > > > > values
> > > > > > > > > > but
> > > > > > > > > > > not
> > > > > > > > > > > > > > frequency of dist (they assume dist is a
> > > > probability
> > > > > > > > > > statement
> > > > > > > > > > > and
> > > > > > > > > > > > > > not a constant or series of 
constants).... to
> > be
> > > > > > > precise
> > > > > > > > > > about
> > > > > > > > > > > it
> > > > > > > > > > > > > > they assume that if it can happen it will.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > So, stationarity is the issue.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > So many people are confusing variance with
> > non-
> > > > > > > > > > stationarity ....
> > > > > > > > > > > > > they
> > > > > > > > > > > > > > are being fooled by randomness e.g.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > we know that the trial records of fair 
coin
> > > > tosses
> > > > > are
> > > > > > > > > > > stationary
> > > > > > > > > > > > > AND
> > > > > > > > > > > > > > they have a surprising range of outcomes
> > > > > (variance) ...
> > > > > > > > > this
> > > > > > > > > > is
> > > > > > > > > > > > > very
> > > > > > > > > > > > > > easy to see if simulated and expressed as
> > equity
> > > > > > > outcomes.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Therefore, in trading, we can, at the 
least
> > > > expect a
> > > > > > > > > > tremendous
> > > > > > > > > > > > > > amount of variance ... no less than what 
can
> > be
> > > > > > > expected
> > > > > > > > > from
> > > > > > > > > > a
> > > > > > > > > > > > > coin
> > > > > > > > > > > > > > toss experiment ... this variance can be
> > estimated
> > > > > > > using
> > > > > > > > > > several
> > > > > > > > > > > > > > methods, simulation being one easy, push 
the
> > > > > computer
> > > > > > > > > button
> > > > > > > > > > and
> > > > > > > > > > > > > look
> > > > > > > > > > > > > > at the graph method.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > So, the value of the simulation is in 
training
> > > > the
> > > > > mind
> > > > > > > > to
> > > > > > > > > > > accept
> > > > > > > > > > > > > > variance and mentally prepare for the 
worst
> > case
> > > > > losses.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > However, it doesn't matter how we design 
our
> > > > > systems we
> > > > > > > > can
> > > > > > > > > > not
> > > > > > > > > > > do
> > > > > > > > > > > > > > anything about stopping non-stationarity.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Our system will get wiped out in OOS if 
it is
> > not
> > > > > > > robust
> > > > > > > > OR
> > > > > > > > > > if
> > > > > > > > > > > the
> > > > > > > > > > > > > > market changes.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > If our system is robust it will still get
> > wiped
> > > > out
> > > > > if
> > > > > > > > the
> > > > > > > > > > > market
> > > > > > > > > > > > > > changes.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > However, IMO, non-stationarity is not, or 
need
> > > > not
> > > > > be,
> > > > > > > as
> > > > > > > > > > > pervasive
> > > > > > > > > > > > > > in trading as we think.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > As I have said in the past, and already in
> > this
> > > > > > > post ...
> > > > > > > > > many
> > > > > > > > > > > > > traders
> > > > > > > > > > > > > > are slayed by the innocuous looking Black
> > Swan,
> > > > > because
> > > > > > > of
> > > > > > > > > > > > > ignorance
> > > > > > > > > > > > > > about its behaviours.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Also, we are very lucky, in trading, to be
> > able
> > > > to
> > > > > have
> > > > > > > > some
> > > > > > > > > > > > > control
> > > > > > > > > > > > > > over our dataset i.e. our sample space is
> > bounded
> > > > by
> > > > > > > our
> > > > > > > > > > stops
> > > > > > > > > > > and
> > > > > > > > > > > > > > other inherent factors in the design.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Example:
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > If we have a stop in place then we are
> > reasonably
> > > > > > > > unlikely
> > > > > > > > > to
> > > > > > > > > > > > > > experience losses beyond the stop +
> > commission +
> > > > > > > > > > slippage ....
> > > > > > > > > > > when
> > > > > > > > > > > > > a
> > > > > > > > > > > > > > stop failure does occur it is very 
infrequent
> > and
> > > > > not
> > > > > > > > > > > necessarily
> > > > > > > > > > > > > > career destroying.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > When we have a profit stop in place we can
> > expect
> > > > > to at
> > > > > > > > > least
> > > > > > > > > > > get
> > > > > > > > > > > > > the
> > > > > > > > > > > > > > stop OR BETTER.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > We can also, in some circumstances, buy a
> > > > guaranteed
> > > > > > > stop
> > > > > > > > > > loss.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > In summary:
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Because, as traders, we are statistically
> > lucky,
> > > > we
> > > > > can
> > > > > > > > > > choose,
> > > > > > > > > > > to
> > > > > > > > > > > > > > some extent, which marbles to put in the 
bin.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > We can absolutely limit the worst case,
> > ensure we
> > > > > get
> > > > > > > at
> > > > > > > > > > least
> > > > > > > > > > > the
> > > > > > > > > > > > > > best case and then take everything in 
between
> > that
> > > > > > > comes
> > > > > > > > > > along.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Since the boundaries are limited, the 
range of
> > > > > possible
> > > > > > > > > > values
> > > > > > > > > > > on
> > > > > > > > > > > > > the
> > > > > > > > > > > > > > balls is finite and will always be 
normally
> > > > > > > distributed,
> > > > > > > > > when
> > > > > > > > > > > > > > expressed as possible mean P & L (central
> > limit
> > > > > > > > > > theoreom).....
> > > > > > > > > > > the
> > > > > > > > > > > > > > staging post on the trail towards possible
> > equity
> > > > > > > > outcomes.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > I think under those circumstances that the
> > balls
> > > > in
> > > > > the
> > > > > > > > > > bucket,
> > > > > > > > > > > > > > collected over a long sample, are a pretty
> > fair
> > > > > > > > > > representation
> > > > > > > > > > > of
> > > > > > > > > > > > > > what we can expect in the future.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > If they are not then we only have 
ourselves to
> > > > blame
> > > > > > > for
> > > > > > > > > our
> > > > > > > > > > > poor
> > > > > > > > > > > > > > system design.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Nothing anyone can do, can put an end to
> > > > stockmarket
> > > > > > > non-
> > > > > > > > > > > > > stationarity
> > > > > > > > > > > > > > but the challenge for the trader is to 
find
> > ways
> > > > to
> > > > > > > > either
> > > > > > > > > > > absorb
> > > > > > > > > > > > > it
> > > > > > > > > > > > > > or anticipate it.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > One important point was absent from your 
post.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Kelly and Vince et al have proved 
conclusively
> > > > that
> > > > > > > > staking
> > > > > > > > > > > > > directly
> > > > > > > > > > > > > > and remarkably affects outcomes.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Based on that fact I can't understand why 
you,
> > > > and
> > > > > many
> > > > > > > > > other
> > > > > > > > > > > > > > commentators, continue to draw inferences 
from
> > > > > > > backtests
> > > > > > > > > that
> > > > > > > > > > > > > include
> > > > > > > > > > > > > > a limited range of portfolio 
allocations ...
> > > > either
> > > > > > > don't
> > > > > > > > > > > involve
> > > > > > > > > > > > > eq
> > > > > > > > > > > > > > at all OR test across all possible eq
> > allocations.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > (if you do opt for the latter choice 
wouldn't
> > it
> > > > be
> > > > > > > > smarter
> > > > > > > > > > to
> > > > > > > > > > > do
> > > > > > > > > > > > > > that using the short mathematical solution
> > rather
> > > > > than
> > > > > > > > the
> > > > > > > > > > long
> > > > > > > > > > > > > > massive optimisation approach?).
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > The babblers epilogue:
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > I guess it is appropriate that an informal
> > book
> > > > > should
> > > > > > > > have
> > > > > > > > > an
> > > > > > > > > > > > > > informal ending!
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > "Always look on the bright side of 
life" ...
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > ... from the life of Brian :-)
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > --- In 
amibroker@xxxxxxxxxxxxxxx<amibroker%40yahoogroups.com>
> > <amibroker%
> > 40yahoogroups.com>
> > > > <amibroker%
> > > > > 40yahoogroups.com>,
> > > > > > > "Howard Bandy"
> > > > > > > > > > <howardbandy@>
> > > > > > > > > > > > > > wrote:
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Greetings all --
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > The posting was originally made by me to
> > Aussie
> > > > > Stock
> > > > > > > > > > Forums
> > > > > > > > > > > on
> > > > > > > > > > > > > > > February 2, 2009. But in light of recent
> > > > > > > discussions,
> > > > > > > > > I'll
> > > > > > > > > > > cross
> > > > > > > > > > > > > > post
> > > > > > > > > > > > > > > it here.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Some of my thoughts on using Monte Carlo
> > > > > techniques
> > > > > > > > with
> > > > > > > > > > > trading
> > > > > > > > > > > > > > systems.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > First, some background.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Monte Carlo analysis is the application 
of
> > > > > repeated
> > > > > > > > random
> > > > > > > > > > > > > sampling
> > > > > > > > > > > > > > > done in order to learn the 
characteristics
> > of
> > > > the
> > > > > > > > process
> > > > > > > > > > > being
> > > > > > > > > > > > > > studied.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Monte Carlo analysis is particularly 
useful
> > when
> > > > > > > closed
> > > > > > > > > form
> > > > > > > > > > > > > > solutions
> > > > > > > > > > > > > > > to the process are not available, or 
are too
> > > > > > > expensive
> > > > > > > > to
> > > > > > > > > > > carry
> > > > > > > > > > > > > out.
> > > > > > > > > > > > > > > Even in cases when a formula or 
algorithm
> > can
> > > > > supply
> > > > > > > the
> > > > > > > > > > > > > information
> > > > > > > > > > > > > > > desired, using Monte Carlo analysis can
> > often
> > > > be
> > > > > used.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Here is an example of Monte Carlo 
analysis.
> > > > Assume
> > > > > > > that
> > > > > > > > a
> > > > > > > > > > > student
> > > > > > > > > > > > > is
> > > > > > > > > > > > > > > unaware of the formula that relates the
> > area of
> > > > a
> > > > > > > > circle
> > > > > > > > > to
> > > > > > > > > > > its
> > > > > > > > > > > > > > > diameter. A Monte Carlo solution is to
> > > > > conceptually
> > > > > > > > draw
> > > > > > > > > a
> > > > > > > > > > > square
> > > > > > > > > > > > > > with
> > > > > > > > > > > > > > > sides each one unit in length on a 
graph,
> > with
> > > > the
> > > > > > > > origin
> > > > > > > > > > at
> > > > > > > > > > > the
> > > > > > > > > > > > > > lower
> > > > > > > > > > > > > > > left corner. The horizontal side goes 
from
> > 0.0
> > > > to
> > > > > 1.0
> > > > > > > > > along
> > > > > > > > > > > the x-
> > > > > > > > > > > > > > axis
> > > > > > > > > > > > > > > and the vertical side goes from 0.0 to 
1.0
> > > > along
> > > > > the
> > > > > > > y-
> > > > > > > > > > axis.
> > > > > > > > > > > Draw
> > > > > > > > > > > > > a
> > > > > > > > > > > > > > > circle with a diameter of one unit 
inside
> > the
> > > > > square.
> > > > > > > > The
> > > > > > > > > > > center
> > > > > > > > > > > > > of
> > > > > > > > > > > > > > > the circle will be at coordinates 0.5, 
0.5.
> > The
> > > > > Monte
> > > > > > > > > Carlo
> > > > > > > > > > > > > process
> > > > > > > > > > > > > > to
> > > > > > > > > > > > > > > compute the area of the circle is to
> > generate
> > > > many
> > > > > > > > random
> > > > > > > > > > > points
> > > > > > > > > > > > > > > inside the square (each point a pair of
> > number
> > > > > with
> > > > > > > the
> > > > > > > > > > > values of
> > > > > > > > > > > > > > the
> > > > > > > > > > > > > > > x-coordinate and y-coordinate being 
drawn
> > from a
> > > > > > > uniform
> > > > > > > > > > > > > > distribution
> > > > > > > > > > > > > > > between 0.0 and 0.999999), then count 
the
> > > > number
> > > > > of
> > > > > > > > those
> > > > > > > > > > > points
> > > > > > > > > > > > > > that
> > > > > > > > > > > > > > > are also inside the circle. The ratio
> > between
> > > > the
> > > > > > > > number
> > > > > > > > > of
> > > > > > > > > > > points
> > > > > > > > > > > > > > > inside the circle to the number of 
points
> > drawn
> > > > > gives
> > > > > > > > an
> > > > > > > > > > > estimate
> > > > > > > > > > > > > of
> > > > > > > > > > > > > > > the constant pi. Running this experiment
> > several
> > > > > > > times,
> > > > > > > > > > each
> > > > > > > > > > > using
> > > > > > > > > > > > > > > many random points, allows application 
of
> > > > > statistical
> > > > > > > > > > analysis
> > > > > > > > > > > > > > > techniques to estimate the value of pi 
to
> > > > within
> > > > > some
> > > > > > > > > > probable
> > > > > > > > > > > > > > > uncertainty. The process being studied 
in
> > that
> > > > > > > example
> > > > > > > > is
> > > > > > > > > > > > > > stationary.
> > > > > > > > > > > > > > > The relationship between the area of the
> > circle
> > > > > and
> > > > > > > the
> > > > > > > > > > area
> > > > > > > > > > > of
> > > > > > > > > > > > > the
> > > > > > > > > > > > > > > square is always the same.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > When we are developing trading systems, 
the
> > > > > ultimate
> > > > > > > > > > question
> > > > > > > > > > > we
> > > > > > > > > > > > > are
> > > > > > > > > > > > > > > most often asking is "What is the future
> > > > > performance
> > > > > > > of
> > > > > > > > > this
> > > > > > > > > > > > > trading
> > > > > > > > > > > > > > > system?" Recall that the measure of
> > goodness of
> > > > a
> > > > > > > > trading
> > > > > > > > > > > system
> > > > > > > > > > > > > is
> > > > > > > > > > > > > > > your own personal (or corporate) choice.
> > Some
> > > > > people
> > > > > > > > want
> > > > > > > > > > > highest
> > > > > > > > > > > > > > > compounded annual return with little 
regard
> > for
> > > > > > > > drawdown.
> > > > > > > > > > > Others
> > > > > > > > > > > > > > value
> > > > > > > > > > > > > > > systems that have low drawdown, or
> > infrequent
> > > > > > > trading,
> > > > > > > > or
> > > > > > > > > > > whatever
> > > > > > > > > > > > > > > else may be important. But, in all 
cases,
> > the
> > > > > goal is
> > > > > > > > to
> > > > > > > > > > have
> > > > > > > > > > > the
> > ...
> >
> > [Message clipped]
>




------------------------------------

**** IMPORTANT PLEASE READ ****
This group is for the discussion between users only.
This is *NOT* technical support channel.

TO GET TECHNICAL SUPPORT send an e-mail directly to 
SUPPORT {at} amibroker.com

TO SUBMIT SUGGESTIONS please use FEEDBACK CENTER at
http://www.amibroker.com/feedback/
(submissions sent via other channels won't be considered)

For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
http://www.amibroker.com/devlog/

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/amibroker/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/amibroker/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:amibroker-digest@xxxxxxxxxxxxxxx 
    mailto:amibroker-fullfeatured@xxxxxxxxxxxxxxx

<*> To unsubscribe from this group, send an email to:
    amibroker-unsubscribe@xxxxxxxxxxxxxxx

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/