|
PureBytes Links
Trading Reference Links
|
Hi Walter
On Sat, 29 Jan 2000 09:02:28 -0500, you wrote:
> ... fat tails are where the money is, therefore leptokurtosis reduction
>removes the money.
This, imo, is not true as a general statement, and I'll come back to
this later.
>The search for market inefficiencies is the search for
>the money.
This is ok for me, but has not to do necessarily anything with
leptokurtosis.
>One possible approach to take account of sets of outliers may now be to use
>the methods described in the previous chapter to remove the impact of these
>aberrant data in order to have a clearer view on the pattern of the
>underlying time series.
So this is a well recognized way to go, especially if I look at high
volatile markets (like options), where even the detection of a rather
small systematical main-stream "market inefficiency" will lead to
reasonable profit, even if a trading system can only make e.g. 1% of
the profit from MetaStock's "Maximum Profit System".
>However, in empirical finance we tend to be less interested in the level of
>the asset price or stock market index since it is widely assumed that such
>time series can be best described as random walks.
Imo, this is a more traditional view (like e.g. the „Efficient Market
Hypothesis“) and is not true in general.
To give an example: First numerical tests that I did on weekly data of
the german main index dax showed, that there is definitely an
attractor with about 10 dimensions, and that the phase space can be
represented with about 16 parameters. This is in strong disagreement
with "random walk" (which I also tested for comparison).
Ergo: There is plenty of room for TA-work directed to systematical
"market inefficiencies", as long as enough people think e.g. the dax
to be a "random walk" function.
(BTW: Are there any comparable results from complexity analysis
around?)
>Therefore, another
>approach is to exploit the fact that outliers emerge in clusters by trying
>to construct a time series model for the outliers themselves.
Concentrating on outliers may be another kind of "market inefficiency"
to look at for the construction of trading systems, especially if I
had to deal with low volatile markets. But because of the nature of
outliers, the statistical basis normally would be much more smaller,
compared to the basis for main-stream "market inefficiencies", leading
to corresponding problems in terms of generalization and consistency
of profits.
In short: Concentrating on outliers & tails is more like poker, and
working on main-stream "market inefficiencies" is more like
"Schafkopfen" (a bavarian version of skat).
Therefore, for me it makes much more sense to concentrate on
main-stream "market inefficiencies", because there are lots of
high-volatile markets, lots of people believing in "random walk
prices", and the generalization problem can be worked-out much better
on a broad statistical basis.
So the objective in system tests could be to
=> maximize profit,
=> minimizes the tail-area of the probability distribution of losses,
=> etc.
This way money is not at all addressed by leptokurtosis (of the loss
distribution) but rather by the integrals of the distributions of wins
and losses. That is,
> ... fat tails are where the money is, therefore leptokurtosis reduction
>removes the money.
is not valid in general, if the systems deal with main-stream "market
inefficiencies".
mfg rudolf stricker
| Disclaimer: The views of this user are strictly his own.
|