Usefulness of Monte Carlo

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
Post Reply
Roundtable Knight
Roundtable Knight
Posts: 1429
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA

Usefulness of Monte Carlo

Post by LeviF » Fri Feb 12, 2010 10:32 am

For some reason I've been pondering the usefulness of the MC simulation lately. It seems to me that if a person developed the ultimate robust system that has a realistic jagged equity curve with lots of long and deep drawdowns, the MC stats for that system will look terrible, but the system will likely perform as tested.

On the other hand, the person that develops the ultimate curve fit system with a straight line equity curve to the moon will have great MC stats, but the system that will not hold up in the future.

So whats the point/value?

As an alternative, I wonder if using monte carlo process to create synthetic data from real data would be useful? You could take fairly large slices across the same time frame for each instrument in your portfolio (to retain both serial & intermarket correlations) and then chain the data together in the same order for each instrument for 1,000 years. Then you can run your system on that to see how it holds up on "out of sample" data. This could be interesting...although fake data is not ideal...

I know these topics have been discussed in the past, i'm just rambling...

Paul King
Roundtable Knight
Roundtable Knight
Posts: 207
Joined: Mon Feb 23, 2004 9:13 am
Location: Vermont, USA

Post by Paul King » Sat Feb 13, 2010 8:53 am

"straight line equity curve to the moon will have great MC stats,"

This depends on what your personal definition of "great" looks like.

If I see a MC simulation with very narrow range of possible equity curves, virtually no drawdowns, and an actual equity curve that's "a straight line equity curve to the moon" I just try to find out where the risk is "hidden", or how it's been deferred (or simply ignore it as a scam or using data not available at the time).

For example, selling 1% worth of strangles per day/week/month gives a very nice looking equity curve (until it doesn't ;-)


Roundtable Knight
Roundtable Knight
Posts: 218
Joined: Wed Aug 09, 2006 4:01 pm

Post by nodoodahs » Mon Feb 15, 2010 5:23 pm

OTOH selling fully-capitalized just OTM SPX puts looks VERY good if you don't have 1987 and 2008 in the backtest ... and if you DO have 1987 and 2008 in the backtest, it STILL looks PRETTY DARN GOOD compared to buy+hold the underlying index from 1986 through 2009.

Sometimes we overstate the risks of optionality ... granted, we can't ignore that skewness and kurtosis blow up a bunch of our simpler metrics and mean we have to search for the hidden risks, but still we should be careful not to throw out babies with bathwater ...

Full Member
Full Member
Posts: 18
Joined: Wed Feb 09, 2005 10:01 am
Location: Germany

Monte Carlo Simulation...

Post by zentrader » Fri Mar 19, 2010 2:44 pm


the so-called "Monte Carlo Simulation stress test (= system simulation)" shows a better analyse of possible chances (profits) and risks (drawdowns), because it uses bigger numbers (but always based on historical system test results and the market conditions of this data!!!).

If you simulate or generate new test data using MCS techniques (= data simulation) you can test your system also with this data, ride new system tests and even new mcs based stress tests. If your system survives also in the most of these (mathematically infinite) scenarios you have a very robust system setup.

Methodically it's most you can do - but how always in life: there's a remaining rest risk (the black swan etc.)... :-)

I've written a book and software concerning this and described this methods in an article in "Trader's" here:


Post Reply