|
PureBytes Links
Trading Reference Links
|
Hi Veysel --
I am not trying to be a smart aleck -- but my last message explained the conditions under which trading the equity curve might work, and now it is your job to try some experiments using your own data and your own trading systems..
Changing the time frame might help. Whatever you do while backtesting, be certain that you will be able to make the same calculations in real time. And, as always, follow good validation procedures -- divide the data into in-sample and out-of-sample periods, backtest over the in-sample data, evaluate using the out-of-sample results.
Thanks,
Howard
On 7/21/07, veysel serbes <vserbe@xxxxxxxxx> wrote:
Howard,
Thank you very much for your explanations. I have a
trading system on 10 min YM futures. It has such a
shape 'a period of earnings followed by a short period
of loosings'. Then it means negatively corelated.
Then ý will use crossover sytems. But my questions is
if ý comppress the equity curce to an upper time frame
, would it be better?
--- Howard B <howardbandy@xxxxxxxxx> wrote:
> Hi Veysel --
>
> The technique you are describing is often called
> "trading the equity
> curve." In order for it to work, there must be
> either a positive
> correlation or a negative correlation between
> sequential points in the
> equity series that you will be using.
>
> Assume that you are trading either intraday or
> end-of-day, computing your
> account equity daily, and want to use an equity
> curve filter based on the
> daily equity.
>
> Computing the equity curve is easy: e = equity();
>
> If daily equity data points are positively
> correlated, then winning days
> tend to be followed by winning days, and the
> category of techniques that
> will work are trend following techniques -- for
> example, the moving average
> crossover you suggest.
>
> If daily equity data points are negatively
> correlated, then winning days
> tend to be followed by losing days, and the category
> of techniques that will
> work are mean reversion techniques -- for example,
> RSI.
>
> If there is little or no correlation between daily
> equity data points, then
> trading the equity curve will not work.
>
> There are other ways to interpret equity than the
> daily equity. For
> example, you could measure the equity only after a
> trade has closed. In
> this case, the series of equity data points would
> have an entry for each
> closed trade and the filter would allow or block
> entry into the next trade,
> but it would not signal an early exit from an open
> trade.
>
> I hope this helps.
>
> Thanks,
> Howard
> www.quantitativetradingsystems.com
>
>
> On 7/21/07, veysel serbes <vserbe@xxxxxxxxx> wrote:
> >
> >
> > I'm thinking about writing a simple MA crossover
> > systems on equity curve drived from a aystem. I'm
> > thinking that this will help me to avoid big
> drawdowns
> > and to wait the market to come back to a mode
> which
> > favors my system.
> >
> > Does any body work on this? How can ý do that?
> >
> >
>
__________________________________________________________
> > Pinpoint customers who are looking for what you
> sell.
> > http://searchmarketing.yahoo.com/
> >
> >
>
__________________________________________________________
Be a better Heartthrob. Get better relationship answers from someone who knows. Yahoo! Answers - Check it out.
http://answers.yahoo.com/dir/?link=list&sid=396545433
__._,_.___
Please note that this group is for discussion between users only.
To get support from AmiBroker please send an e-mail directly to
SUPPORT {at} amibroker.com
For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
http://www.amibroker.com/devlog/
For other support material please check also:
http://www.amibroker.com/support.html
![]()
SPONSORED LINKS
__,_._,___
|