Re: Question About MAE/MFE?

["Ryugan Mizuta" <trend456@xxxxxxxxxxx>]

Thanks for the info.  Do u have any suggestion on how I can create better 
targets and stop?

Thanks,again.

Ryugan M.

From: Alex Matulich <alex@xxxxxxxxxxxxxx>
To: Ryugan Mizuta <trend456@xxxxxxxxxxx>
CC: omega-list@xxxxxxxxxx
Subject: Re: Question About MAE/MFE?
Date: Mon, 28 Jul 2003 09:52:23 -0700 (PDT)

Ryugan M wrote:
>I'm trying to create better stops and profit targets for my systems and
>decided to use MAE and MFE. I'm looking for a EL code or anything that 
links
>to tradestation so that I can automate this process.  If anybody has any
>suggestions please send me a reply.

I have some code that outputs Maximum Adverse Excursion (MAE), at
http://unicorn.us.com/trading/el.html - click on _SystemQuality.

You may find that MAE isn't as useful as the literature says.  It
*sounds* logical -- measure the maximum adverse excursion of your
winning trades, and use that to set your initial stoploss.  In
theory this lets your winners win and stops out the losers early.

In practice, I noticed that using MAE to set initial stops made
performance worse.  It doesn't work to measure the MAE of your
winners and then set your initial stop at the MAE point.  When I did
this, the losers lost more - in many cases I had losers that would
recover a bit before being stopped out by other means, but stopping
them out at MAE meant they lost more.

Setting the initial stop narrower so that 95% of the MAE winners
wouldn't get stopped out, just means that 5% of winning trades
became losing trades, reducing the expectancy of my strategy.

Eventually I gave up on doing anything with MAE.

--
  ,|___    Alex Matulich -- alex@xxxxxxxxxxxxxx
 // +__>   Director of Research and Development
 //  \
 // __)    Unicorn Research Corporation -- http://unicorn.us.com