|
PureBytes Links
Trading Reference Links
|
Bill, I'm not sure what you mean by "falsifiable hypothesis", but would
the Superbowl Indicator qualify?
Some basic background details at:
http://www.snopes.com/business/bank/superbowl.asp
Based purely on the historical "accuracy" rate this indicator is an
excellent one, correct 80% of the time. Clearly there's no causal
relationship between who wins the Superbowl and what the stock market
does, but how could its "success" as a market indicator be disproven
using only the historical data we used in testing?
That's the fundamental problem with testing only using numbers and not
thinking about the "whys."
Now...what's the difference between the Superbowl Indicator and a
200-day moving average? If a stock crosses above a 200-day moving
average it's widely assumed that this means the stock is in an uptrend.
However, a 200-day ma applied to randomly-generated numbers can also
show an "uptrend" when charted. Why is there a causal relationship with
one but not the other? Who says?:) Using only the historical data
employed by testing, prove it.:)))
So, here we are...there's no way to either prove or disprove if the
200-day ma is a legitimate indicator or not. We know there's no causal
relationship there, the stock isn't going up because the 200-day ma is
going up, it's the other way around.
So what DOES make a stock going up? Whatever that "X" factor is, why
not focus directly on that instead of something that might just be a
meaningless mathematical coincidence?
Just trying to stimulate some thought about what we really know and what
we don't really know. I don't have all the answers, either, I'm still
trying to get rid of all the stuff I "know" that's wrong.:)
Luck,
Sebastian
--- In amibroker@xxxxxxxxxxxxxxx, "wavemechanic" <fimdot@xxx> wrote:
>
> Brian:
>
> I don't recall being promised anything (but then my short term memory
is shot). I did say to Sebastian that providing a falsifiable
hypothesis would provide a good "chew." Be that as it may, your
hypothesis is that Mondays are different from the average behavior of
all other days. OK, so now what? Off hand it is not clear to me that
the hypothesis provides much "chewing" exercise. How should we use this
information or is it just an academic exercise?
>
> Bill
>
>
> ----- Original Message -----
> From: "brian.z123" brian.z123@xxx
> To: amibroker@xxxxxxxxxxxxxxx
> Sent: Thursday, October 26, 2006 10:59 PM
> Subject: [amibroker] Re: OT: Statistics
>
>
> Hypothesis for Bill.
> I promised Bill *The Wave Mechanic* that I would provide the forum
> with something more substantial than the *candy floss* of philosophy
> and psychology.
> Honouring my promise here is my attempt.
>
> It is an example of proceeding from a subjective view of the markets
> (theory, intuition, guess, hunch etc) to an objectively verified
> hypothesis that could form the basis of a trading plan.
> It is intended as an introductory topic to trading.
>
> I dedicate this post to all mathematicians past and present.
> God knows why he/she didn't include me in that group; it appears
> he/she had other plans for me.
>
>
> *******************************************************************
> General Theory (of market relativity) No1.
>
> The behaviour of markets in the major western cultures is similar
> (USA, Australia, UK & others?)
> The markets are moved by institutional investors who control the
> major portion of the money.
> Institutional behaviour has characteristics that can be utilised by
> freelance traders.
> Institutional investors are people and will exhibit some behaviours
> typical of people in general.
> Based on personal observation of institutions, people and people in
> institutions I expect markets to exhibit some of the following
> behaviours to some extent:
> ( I don't comment on other markets because I haven't lived in those
> countries and don't understand the psyche the way I understand the
> psyche of the above group. That doesn't place any relative value
> judgement on other cultures).
>
> Institutions have budget cycles (weekly, monthly, quarterly, yearly?)
> Institutions are bound by rules of governance and obligations.
> It is not in the interests of institutional players to stick their
> heads up out of the trench.
> Institutions set employees targets which are benchmarked against
> industry norms.
> People have short memories.
> People *work* from Mon-Fri, 9-5 .
> People get Mondayitis and Fridayitis.
>
> There are many more similar observations.
> I have only provided a few as examples.
>
> ********************************************************************
>
> Hypothesis derived from General Theory No1.
>
> The markets will behave differently on Mondays compared to average.
>
> ********************************************************************
>
> Verification of the Hypothesis.
>
> The following is summarised from R.W Colby's *Encyclopedia of
> Technical Market Indicators*, published by McGraw-Hill.
> The original is from the writings of Arthur A. Merrill author of
> *Behaviour of Prices on Wall St*.
> It is presented for educational purposes.
>
> The Chi-squared test ( a statistical method) is used to tell us how
> reliable an indicator is and if the patterns exhibited by data could
> have been produced by chance.
> The source of the data that the test is performed on is not given,
> which means that the results are not independently verifiable by the
> forum.
> On that basis it is only provided to demonstrate the principles.
>
> Trading from 1952 to 1983:
>
> the number of Mondays when the market rose was 669
> versus Monday falls 865
> (1534 in total).
> Average for all trading days = = 52.1% updays.
> Expected rises for Monday = = 52.1% x 1534 = = 799.
> Expected downs for Mondays = = 47.9% x 1534 = = 735.
>
> Using the absolute of the (actual - expected) values for up/down
> days the Chi-squared test (with Yates correction) for statistical
> signifance returns the value 43.81;
>
> (((abs(a1-e1) - 0.5)^2)/e1) + (((abs(a2-e2) - 0.5)^2)/e2)
>
> (((abs(669-799) - 0.5)^2)/799) + (((abs(865-735) - 0.5)^2)/735) = =
> 43.81
>
> Based on the Chi-test score there is less than one chance in 1000
> that the observed outcome for Mondays was due to chance alone.
>
> ********************************************************************
>
> Please note!
>
> I can't comment on the veracity of the Chi-squared test.
> I will have to leave that to the mathematicians and rightly so.
> I provided the example only to stimulate debate or interest in
> statistics as a powerful trading tool.
>
> The general theory of the market is my own personal view.
> I have not verified the above results myself nor have I attempted to
> develop a trading system based on a Monday trading cycle, or any
> other calender cycle for that matter.
> I have, however, successfully developed and tested a trading system
> based on another hypothesis that was derived from the above
> GeneralTheory.
>
>
>
> BrianB2......<:-)......(the last of the coneheads?).
>
>
>
>
> --- In amibroker@xxxxxxxxxxxxxxx, "brian.z123" brian.z123@
> wrote:
> >
> > Statistics for traders.
> > Can anyone recommend a book on statistics written specifically for
> > traders or that applies statistical methods to trading examples?
> > I am looking for an author who has done a good job on the subject.
> > Even if it is only a section of a book that would do provided it
> > goes beyond a superficial treatment of the subject.
> >
> > For anyone interested here is a link to a very good introduction
> or
> > refresher for statistics.
> > The HTML *book* takes your from 0-50kph in approx 100 pages.
> > Please note; the site does contain a lot of advertisements but it
> is
> > also a mini portal for stats and it does have links to free
> > statistical stuff and free tools.
> >
> > Outside of writing indicators I find statistics to be one of the
> few
> > maths disciplines that has a high degree of relevance to trading.
> >
> > http://davidmlane.com/hyperstat/index.html
> >
> > BrianB2.
> >
>
>
>
>
>
> Please note that this group is for discussion between users only.
>
> To get support from AmiBroker please send an e-mail directly to
> SUPPORT {at} amibroker.com
>
> For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
> http://www.amibroker.com/devlog/
>
> For other support material please check also:
> http://www.amibroker.com/support.html
>
> Yahoo! Groups Links
>
>
>
>
>
>
>
> --
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.1.408 / Virus Database: 268.13.14/501 - Release Date:
10/26/2006
>
Content-Description: "AVG certification"
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.409 / Virus Database: 268.13.27/517 - Release Date: 11/3/2006
|