Sunday, March 30, 2008

Portfolio Update 28/03/2008



No entries to take next week.


Maximum drawdown on closed equity has reached 27.25% which greater than the maximum drawdown seen during testing.


I have made a decision to keep the system open until it reaches 2*max.DD which was seen in testing. This would be 26%*2 = 52%.


I have been reminded that backtesting is not an exact science, and some allowance needs to be made for the system as the biggest maximum drawdown may well be in the future.


The pain realised during this drawdown, I have concluded, will be less than the pain realised if I was to close the system now right before the market takes off again.


And if the system does reach the newly defined shut off point, I will live to trade another day.

4 comments:

Cameron said...

Hi Nizar,

At risk of sounding like somebody else...do you know why your system is outside the bounds of testing? Or more to the point, do you know why your system is not managing the trading conditions of the last 6 months?

ASX.G

Nizar said...

Hi ASX.G

Thanks for your comments.

My system is not designed to make money in bearmarkets.

The market volatility that my system is facing currently is beyond that seen in testing, and as such the maximum drawdown has surpassed the maximum drawdown that would have been seen during the 2000-2002 bear market.

Nizar.

Cameron said...

Hi Again,

Did your testing include June 06, or March 07, or August 07 corrections?

If so, how did it hold up during these periods?

ASX.G

Nizar said...

Hi ASX.G,

Yes my testing did include these periods, but I did not simulate a situation where I would start the portfolio at the peaks before these corrections, and let me explain why.

The system, being long term trend following in its methodology, is not designed to, nor is it expected to, make money over the short term.

Because of this, I made a decision to conduct all backtests over a period of a minimum of 2 years. As such, the corrections you mention above were traded through and they did not adversely affect the results.

I did though, start a test on 01-07-2001 which is the peak before the last bearmarket we had. This is where the 25.6% figure max.DD came from.

Nizar.