|
PureBytes Links
Trading Reference Links
|
Hi Lee,
Here is a link to one of the best overviews of position sizing and
capital risk that I have seen. It is written by a Ph.D. (Michael
Bryant) that has found a single formula that unifies all of the
standard methods that we read about (Fixed Fraction, Fixed Ratio,
etc.).
http://www.breakoutfutures.com/Downloads/TradeRiskTalk.ppt
You will need Powerpoint to view it. If you don't have Powerpoint,
you can download a free copy of Powerpoint viewer from Microsoft's
website.
Bryant has a web site called Breakout Futures
http://www.breakoutfutures.com/
If you look under the "Free Donwloads" section, you will find lots of
information on position sizing and his trading methods.
Bryant has presented this work (and other versions with more
extensive explanations) at various meetings that I have attended.
The files are too big to share through the EquisMetaStock site. If
you want copies sent, then let me know. Warning: extensive math in
these presentations.
Ross
--- In equismetastock@xxxxxxxxxxxxxxx, "leeontherun"
<leeontherun@xxx> wrote:
>
> POSITION SIZE CALCULATION & CAPITAL RISK
>
> There seems to be a lot on formula generation but one thing I
learnt
> after a few years of trading is that your formulas and backtest
work
> can pretty much be thrown out the window if you don't take the time
> to size up your risk especially if you are heavily margined.
> Personally, with CFD's I can margin 100 times my intial equity, so
> there is huge risk involved.
>
> I'd like to share with you some results of research and in return
> hope for other's research on position sizing or articles they have
> read and can submit. It is worth taking the time to post this
> information.
>
> In the Equismetastock yahoo group 'Photos' you will find a folder
> called 'Position Sizing' that this post refers to. This folder has
> several graphs on position size calculations. I'd advise you visit
> the photos section now and save the pictures to your computer.
>
> My research involves using a % of your capital to trade in place of
a
> fixed position size. That means that as you take winners and your
> capital grows, so does your position size and in effect gives you
> exponential growth. The reverse of this is if you were to take some
> losers first then your % of capital to trade size will be getting
> smaller, but in the long run you should be able to pull yourself
out
> if your backtested plan shows larger number of winners than losers.
>
> This backtested plan showed the largest number of consecutive
losers
> was given to be '4' with the average loss size shown on the left
hand
> side of the results chart depending on what % of capital you trade.
>
> Now here is the trick!
>
> Lets assume we hit the worst case scenario and took '4' consecutive
> losers first before we had a regular pattern of winners and losers.
>
> When graphing this information, if you look at the chart of 4
> consecutive losers it shows you a sweet spot of what % of capital
> would be best used before it starts to have a poorer efficiency
i.e.
> you ideally would like to keep on the left hand side of the curve
> before the peak. So in this example 10% or maybe even 12.5% of
> capital should be traded.
>
> Now say you wanted to play it safe and use a fudge factor that we
may
> get 5 consecutive intitial losers or maybe even 6, then you just
need
> to look at the respective charts.
>
> This is so far working great for my trading.
>
> Lee.
>
------------------------------------
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/equismetastock/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/equismetastock/join
(Yahoo! ID required)
<*> To change settings via email:
mailto:equismetastock-digest@xxxxxxxxxxxxxxx
mailto:equismetastock-fullfeatured@xxxxxxxxxxxxxxx
<*> To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|