|
PureBytes Links
Trading Reference Links
|
This keeps turning up on one guy's email...
"> > Felix,
> >
> > This is a classic example of academic research
> > done by someone who, pardon the vulgarity, doesn't
> > know dick about trading.
> >
> > Walt
> >"
As an academic in financial markets, former local and active trader, I
suggest you do a little more homework before making these kinds of
remarks. Sadly, that sort of anti-intellectual sentiment is the reason
why so few 'pros' don't bother to read any of the academic journals and
keep themselves happy with what are little more than unrefereed trade
journals.
The burden of proof (eg in Efficient Market Hypothesis testing)is a lot
higher than you might think and poorly crafted research is quickly
dispensed with. Unfortunately a lot of research put out in unrefereed
works rely on a large amount of faith in the testing methods and
integrity of the promoter (and who knows if it's just another get rich
quick merchant providing unsupportable and untestable hypotheses).
As I am currently completing a PhD on Tech An in futures (after
producing a Masters thesis on the same topic), I can assure you that
when you have to really critically evaluate people's evidence - you
realise how much in the unrefereed sphere is simply unsustainable.
For example, in Balsar, N. (1992) 'Money Management Strategies for
Futures Traders', Wiley, USA. the comment is made under the headng 'Head
& Shoulders Formation' that "Perhaps the most reliable of all reversal
patterns, this formation can occur either as a head-and-shoulders top
or..." (p25). These sorts of comments appear in numerous works, though
no evidence is provided. How reliable? How many times was it witnessed
in a test period? What were the drawdowns? What sort of funds are
required to make and sustain the trade so that a percentage annual
return can be estimated? Does this allow for slippage? What the...
This is not meant as a criticism of Balsara per se, but it is indicative
of the sort of claims that are howled down in an academic environment
because no evidence is provided, no clear methodology is defined...
nothing. I agree that some of my colleagues in the University sector
lack practical experience as traders, but I'm pretty certain that the
amount of Tech Analysts publishing in this area are a damned sight worse
when it comes to constructing sound, rigourous and replicable testing
methodologies. How many times have you picked up a book with lots of buy
(sell) signals, but no attempt to define when the trade should be
exited? For example, the double bottom buy signal, OK I'm long, now
what? The results of such testing are largely worthless because they
could not be verified using the methods suggested by the author.
Perhaps the standard of proof required on all sides should be raised so
that at last we can actually get somewhere with providing EVIDENCE and
have them reported to a standard that anyone would expect before
investing money in anything. If practitioners would improve their game
at reporting evidence, perhaps the divide between academics and
practitioners could be bridged - even if just for a little while.
John
|