[FoRK] Forex opportunities and kwatloos?

Aaron Burt <aaron at bavariati.org> on Fri May 11 18:24:39 PDT 2007

On Fri, May 11, 2007 at 08:09:24PM -0500, Jeff Bone wrote:
> On May 11, 2007, at 7:56 PM, Aaron Burt wrote:
> >Had a control-theory prof who was trying to get his brighter students
> >(i.e. not me) into working on automated trading.
> >It was all just simpleminded digital low-pass filters hand-tuned to
> >yield good returns on historical data.
> Reality behind the back-testing mirage:  slippage kills.

Ayuh, that was my impression.  I asked him about trade completion time
and he just gave me a blank stare.  I'm not sure he'd read much on
technical trading.  Heck, most of those filters introduce too much
delay, even assuming that volatility hasn't changed between last decade
and now.

Was tempted to try back-testing (useful term!) some simple algorithms
based on flagging and 2/3-retrace behaviors.

> >Someday we'll have corporate bylaws written in Perl,
> I am amused by Charlie Stross's idea of Economics 2.0

Caught me fair-and-square.  That's where I stole it from.

More information about the FoRK mailing list