a Fed Reserve board paper confirming t/a.
clearly using stats.
http://www.ny.frb.org/research/epr/00v06n2/0007osle.pdf
Conclusion
This article has examined the predictive power of support and
resistance levels for intraday exchange rates, using technical
signals published by six active market participants from
January 1996 through March 1998. The statistical tests, which
use the bootstrap technique (Efron 1979, 1982), cover support
and resistance levels for three currency pairs: dollar-mark,
dollar-yen, and dollar-pound.
The results indicate that intraday exchange rate trends
were interrupted at published support and resistance levels
substantially more often than would have occurred had the
levels been arbitrarily chosen. This finding is consistent across
all three exchange rates and across all six firms studied. The
predictive power of published support and resistance levels
varies considerably across firms and across exchange rates.
It lasts at least one week. The strength estimates published
with the levels are not meaningful. These results are highly
statistically significant and are robust to alternative
parameterizations.
The predictive power of support and resistance levels has
many possible sources, some of which are discussed in Osler
(2000). Central bank intervention has been cited as a possible
source of the predictive power of other technical trading
strategies (Szakmary and Mathur 1997; LeBaron 1999).
However, central bank intervention seems unlikely to be an
important source of the predictive power of support and
resistance levels since there was no reported intervention for
the mark and the pound during the sample period. Other
possible explanations include clustered order flow, which
receives support in Osler (2000), and self-fulfilling
prophecies.