October 26, 2010

Information about how to Develop Robust Commodity Trading Systems

By Planet Wealth

p
pWith computers as incredible as they are these days it is simple to optimize a trading system which causes it to seem exceptional, but an optimized system is not a dependable system. Just merely because you are able to train your pc to have 20/20 hindsight does not imply that future performance will be anything like the past./p
pThe main problem with optimizing past performance is that markets shift. A low-volatility market suddenly turns into a high-volatility market. A market inclined to trends becomes a choppy directionless market or, a market which formerly had high leverage becomes a market with small leverage. The list is unlimited./p
pWhat tends to happen is that market X will have a tendency to start behaving like market Y, and market Y will tend to start behaving like market Z. If you have extensively optimized the system to trade market Z, then an individual will be in distress when it starts to trade like market X! This is a issue with many systems, generally stock index systems that tend to be optimized to one market or sector. In spite of their occasional impressive looking results, there is some poison in their mixture./p
pCompare this last situation with one in which the systems model works well with most all of the markets, A thru Z. Now, it will not matter if market Z starts to act like market Y or market A starts to behave like market P. They can transform as many times as they want since the systems structure will be universally robust with most ALL the various markets. Once again, the market qualities can reshuffle numerous times and your system functions like a Swiss Army knife that has demonstrated throughout historical testing it can cope well with most all those situations./p
pThere tend to be a handful of tip offs to an optimized system./p
p1.Unrealistic looking performance/p
p2. Only trades one market or sector well/p
p3. Uses different rules (formulas) for each market/p
p4. Uses different inputs for each market/p
p5. Uses different rules or inputs for getting into buys and sells/p
p6. Does not factor in reasonable transaction costs (slippage amp; commission)/p
p7. Uses money management techniques that do not include market normalization (like single contract performance only)/p
p8. Utilizes static numbers for all markets like a $2000 stop or $5000 profit target (some markets could possibly strike those in an hour and others could take weeks)./p
pAn essential element of a robust system is that it should weight every market evenly. The testing should be done in a way that normalizes the variation between the markets. For instance, natural gas changes an average of a few thousand dollars a day for each contract; but, Eurodollars vary an average of a few hundred dollars a day for each contract. You have to have a method to balance and normalize this kind of difference in testing./p
pThe reason you need to do this is that what if the system satisfies the majority of the above non-optimized rules, but it is trading one natural gas market contract for every Eurodollar contract. The system would look best if it experienced many natural gas winners, but what it natural gas starts to have numerous losing trades and the Eurodollar starts to experience numerous successful trades? Do you reckon a few, hundred dollar winning trades in one Eurodollar contract is going to be sufficient to offset a few THOUSAND dollar losing trades in one natural gas contract?/p
pIf you are trading 20 markets, it is because you want diversification, but if you are trading them all on a one contract basis then you are certainly not diversified. You might have 25% of your account generating 90% of the profits and losses! The problem is that going forward you will be dependent on those markets. It is far superior not to be dependent on any given market in the portfolio. They should all be of the same weight and significance./p
pIn summary a robust system should do the following/p
p1. Trade a portfolio of EVERY commodity market/p
p2. Trade that huge portfolio over a lengthy test period/p
p3.Use the identical rules for every market/p
p4.Use the same input values for every market/p
p5.Include the same logic for entering buys and sells/p
p6. Factor in reasonable transaction costs/p
p7. Normalize the markets for risk/p
/p
pAfter you have done all this, the final step would be to do some walk-forward testing. This means, test and make your system on data up until year 2000 (for instance). Then after doing all the testing, observe how it would have performed from year 2000 until now. This helps avoid numerous advantages of hindsight. These are all things we do in the development of our a href=http://www.traderstech.net target=_blanktrading systems/a./p
pDean Hoffmanbr /DH Trading Systemsbr /www.RelativityTradingSystem.com/p

Topics: Uncategorized | No Comments »

Comments

« | Home | »