|
PureBytes Links
Trading Reference Links
|
Pal,
You description of stationarity vs non-stationarity was nothing short of poetic. I had to read through your e-mail twice, but thanks for taking the time.
I certainly agree it is good to know that we are not dealing with a stationary data series. The big headline I take away from all this discussion is:
1. Beware of over-optimizing.
2. Build in ways of detecting market changes
3. Have circuit breakers on your systems assuming #2 doesn't work.
I'm an Army guy. So, if you can't break it down to some useable rules, it's just talk. After all, we trade where the rubber meets the road, not in the clouds with all the academics.
Regards,
Garypalsanand <palsanand@xxxxxxxxx> wrote:
I agree. I would love to hear your comments on the following:The more you have information, the more you are confident about the outcome. Now the problem: by how much? Common statistical method is based on the steady augmentation of the confidence level, in nonlinear proportion to the number of observations. That is, for an n times increase in the sample size, we increase our knowledge by the square root of n. Suppose I am drawing from an urn containing red and black balls. My confidence level about the relative proportion of red and black balls, after 20 drawings is not twice the one I have after 10 drawings; it is merely multiplied by the square root of 2 (that is, 1.41). Where statistics becomes complicated, and fails us, is when we have distributions that are not symmetric, like the urn above. If there is a very
small probability of finding a red ball in an urn dominated by black ones, then our knowledge about the absence of red balls will increase very slowly – more slowly than at the expected square root of n rate. On the other hand our knowledge of the presence of red balls will dramatically improve once one of them is found. This asymmetry in knowledge is not trivial--it is a central philophical problem for such people as Hume and Karl Popper. I can confirm that an investor trader is a bad investor (if he blows up); but I can never rule out that he may be one. To assess an investor's performance, we either need more astute, and less intuitive, techniques, or we may have to limit our assessments to situations where our judgment is independent of the frequency of these events. But there is even worse news. In some cases, if the incidence of red balls is itself randomly distributed, we will never get to know the
composition of the urn. This is called the problem of stationarity. Think of an urn that is hollow at the bottom. As I am sampling from it, and without my being aware of it, some vicious child is adding balls of one color or another. My inference becomes thus insignificant. I may infer that the red balls represent 50% of the urn while the vicious child, hearing me, would swiftly replace all the red balls with black ones. This makes much of our knowledge derived through statistics quite shaky.The very same effect takes place in the market. We take past history as a single homogeneous sample and believe that we have considerably increased our knowledge of the future from the observation of the sample of the past. What if vicious children were changing the composition of the urn? In other words, what if things have changed?The "science" of econometrics consists of the application of statistics to samples taken at different
periods of time, which we called times series. It is based on studying the times series of economic variables, data, and other matters. Studying the European markets of the 1990s will certainly be of great help to a historian; but what kind of inference can we make now that the structure of the institutions and the markets has changed so much?Stanford economist Mordecai Kurz puts it as follows:The process of structural change (i.e. non-stationarity) in our society is the central building block of its complexity and the root cause of the diversity of beliefs about it. In such a system, the past is not an entirely satisfactory basis for assessment of risks in the future.Practitioners of the "financial engineering" methods measure risks (I just use stops), using the tool of past history as an indication of the future. We will just say that the mere possibility of the distributions not being stationary makes the
entire concept seem like a costly (perhaps very costly) mistake. This leads us to a more fundamental question: the problem of induction which is a different subject.rgds, Pal--- In amibroker@xxxxxxxxxxxxxxx, "Gary A. Serkhoshian" <serkhoshian777@xxxx> wrote:> Ahh-hah ! You see this is where watching the equity curve would tell you that you are bouncing off trees rather than moving through the forest unimpeeded. Also, if you recorded your journey through a large forest rather through a simple patch of trees you would have a better gague on growth patterns of trees and hence your path.> > Taking this a step farther, pine trees grow differently than redwoods. So, you would run into trouble if you took a recording of a journey through a redwood forest, and used it to guide you through a pine forest.> > What people fail to understand is by virtue of
picking some set of parameters we are optimizing. I see people with smug looks on their face when they say they don't believe in optimizing, but are hell-bent on making decisions with a 12,26,9 MACD.> > It's like a social drinker berating an alcoholic. They both drink alcohol, it's just that the social drinker ensures he doesn't wake up in the gutter. Over optimizing is a disease just like over drinking.> > Regards,> Gary> > Joseph Platt <jplatt@xxxx> wrote:> > Lots of talk lately about Optimization, Overoptimization, > Robustivity, Randomness, etc. Regarding optimization here is a copy > of an email I sent to a FastTrack friend of mine recently....don't > know if it makes any sense or not. > **********************************************************************> ******************> You know, most all systems depend on
optimization at some level but > one day when I was in a dreamy mood, (perhaps muscatel induced), this > analogy with regards to optimization popped into my head. Takes a > little imagination....> > Suppose a person decided to take a one mile walk through a well treed > area of woods. He would have no trouble negotiating the path since > the trees are quite visible and he could navigate a path right around > them.> > Suppose further that he had been carrying some kind of an electronic > recorder with him (this is the part that takes a little imagination) > and every step along the one mile stretch was recorded.> > Still feeling energetic he decides to pick up where he left off and > try another one mile hike through the next stretch of woods. He > reasons that it's not necessary to look for an unobstructed path, at > least not as carefully as
he did on the first stretch, because after > all he has a recording of the whole journey and all he has to do is > put it in the "play it again Sam" mode.> > The analogy doesn't have a very happy ending....he got through the > second mile alive but just barely....you would hardly recognize him.> > But to be fair there is a difference between trees that, as far as we > know, grow randomly and the stock market which arguably isn't totally > random as many claim.> **********************************************************************> ******************> I guess my friend got the email OK but his only response was "watch > out for the trees".> > .....Joe Platt> > > > > Yahoo! Groups SponsorADVERTISEMENT> > 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. > > > ---------------------------------> Do you Yahoo!?> Protect your identity with Yahoo! Mail AddressGuardSend BUG REPORTS to bugs@xxxxxxxxxxxxxSend 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.
Do you Yahoo!?
Protect your identity with Yahoo! Mail AddressGuard
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.
|