Automated System Over-Optimization

Page 1

The Inherent Problems with Automated System Over-Optimization

A complimentary publication provided by: www.AutoSystemTrader.com


CFTC RISK DISCLAIMER U.S. GOVERNMENT REQUIRED DISCLAIMER - COMMODITY FUTURES TRADING COMMISSION FUTURES AND OPTIONS TRADING HAS LARGE POTENTIAL REWARDS, BUT ALSO LARGE POTENTIAL RISK. YOU MUST BE AWARE OF THE RISKS AND BE WILLING TO ACCEPT THEM IN ORDER TO INVEST IN THE FUTURES AND OPTIONS MARKETS. DON'T TRADE WITH MONEY YOU CAN'T AFFORD TO LOSE. THIS IS NEITHER A SOLICITATION NOR AN OFFER TO BUY/SELL FUTURES, STOCKS OR OPTIONS ON THE SAME. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFITS OR LOSSES SIMILAR TO THOSE DISCUSSED IN THIS REPORT. THE PAST PERFORMANCE OF ANY TRADING SYSTEM OR METHODOLOGY IS NOT NECESSARILY INDICATIVE OF FUTURE RESULTS. CFTC RULE 4.41 - HYPOTHETICAL OR SIMULATED PERFORMANCE RESULTS HAVE CERTAIN LIMITATIONS. UNLIKE AN ACTUAL PERFORMANCE RECORD, SIMULATED RESULTS DO NOT REPRESENT ACTUAL TRADING. ALSO, SINCE THE TRADES HAVE NOT BEEN EXECUTED, THE RESULTS MAY HAVE UNDEROR-OVER COMPENSATED FOR THE IMPACT, IF ANY, OF CERTAIN MARKET FACTORS, SUCH AS LACK OF LIQUIDITY. SIMULATED TRADING PROGRAMS IN GENERAL ARE ALSO SUBJECT TO THE FACT THAT THEY ARE DESIGNED WITH THE BENEFIT OF HINDSIGHT. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFIT OR LOSSES SIMILAR TO THOSE SHOWN WITHIN THIS REPORT, SUPPORT AND TEXTS. OUR COURSE(S), PRODUCTS AND SERVICES SHOULD BE USED AS LEARNING AIDS ONLY AND SHOULD NOT BE USED TO INVEST REAL MONEY. IF YOU DECIDE TO INVEST REAL MONEY, ALL TRADING DECISIONS SHOULD BE YOUR OWN. THIS SITE IS PROVIDED BY “AUTOMATED SYSTEM TRADER” (AST). PLEASE BE ADVISED THAT AST IS NOT A REGISTERED BROKER, FINANCIAL ADVISOR, NOR IS REGISTERED IN ANY WAY WITH ANY GOVERNMENT REGULATOR AGENCY. AST MAKES THIS INFORMATION, ITS PRODUCTS AND SERVICES AVAILABLE THROUGH THIS WEBSITE UNDER THE FIRST AMENDMENT OF THE UNITES STATES CONSTITUTION. AST HAS MADE GREAT EFFORT TO LIST ALL INFORMATION TO BE VERY ACCURATE. ALL DATA INCLUDED IN THIS WEBSITE IS CONSIDERED HYPOTHETICAL. FUTURES, FOREX, OPTION’S AND ANY TRADING OR INVESTING CONTAINS RISK. ESPECIALLY TRADING IN LEVERAGED VEHICLES SUCH AS FUTURES, FOREX OR RELATED MARKETS. AST PRODUCTS AND SERVICES MAY NOT BE SUITABLE FOR EVERYONE. TRADING SHOULD ONLY BE DONE WITH TRUE RISK CAPITAL. PAST PERFORMANCE IS NOT INDICATIVE OF FUTURE PERFORMANCE.


What is optimization and how does it affect automated trading strategies? Optimization refers to taking an automated system trading program and using a computer to backtest know market data (price action) of a period of time in the past to create the most profitable returns. The problems with computer backtesting or optimization is that when you look forward to trade the system in real time, it may not perform the same as it has in the past because the price action has not been tested. The following pages show automated trading systems that were developed by professional system developers that were over-optimized to create equity curves that did not hold up in real time markets. This publication is designed to demonstrate that the more complex a trading system becomes, the more likely that it can be curve fit. In terms of trading, we believe that automated systems with less input variables are more sustainable in real time trading. Most of the systems that we provide are simple enough to have performed well in out of sample testing and real time trading. All systems that AutoSystemTrader.com provides have gone through no less than 3 months of live trading to ensure that over-optimization has not occurred. AutoSystemTrader.com offers hypothetical trade data for backtested/optimized data, walkforward/out of sample data and real time trade data from the automated strategy signals before slippage and commissions. We provide this only as an indication of how systems have performed. Of course, past performance can never guarantee future results.


The following pages will contain automated trading systems and their respective equity curves that we have tested and are offered to show how over-optimization can and does occur. It is our intention that this information will help the consumer to make a more informed decision before purchasing or trading an automated trading system. First, the backtested equity curve will be presented. This is what the system developer uses to optimize the variables of his or her strategy to produce the highest profit, lowest drawdown, or whatever other parameters that the developer may be trying to identify. Next, when the page is turned, you will see an image that contains the same data along with trades that have occurred after system optimization. On the next page, you will see only the real time trade data for those systems that have been over-optimized. AutoSystemTrader.com offers systems tries to find automated trading systems that have withstood real time trading without breaking down. For this reason, if you see systems on the site that say “Available Soon� it is because we are currently testing these systems to see if they hold up in real time. Typically, we optimize our strategies with at least six months of untested data. We then open the system up to incorporate the most recent six months of untested data. After that has been done, we post the systems up on our site and monitor them for a period of six months before making them available for subscription. We do this only in an attempt to reduce risk of trading an automated system too soon. For more information, please visit us online at http://www.AutoSystemTrader.com


System A Backtest Period - Where Optimization is Performed


System A Backtested Optimization Period with Real Time Data


System A Real Time Trade Data - After Optimization Period


www.AutoSystemTrader.com


System B Backtest Period - Where Optimization is Performed


System B Backtested Optimization Period with Real Time Data


System B Real Time Trade Data - After Optimization Period


www.AutoSystemTrader.com


System C Backtest Period - Where Optimization is Performed


System C Backtested Optimization Period with Real Time Data


System C Real Time Trade Data - After Optimization Period


www.AutoSystemTrader.com


System D Backtest Period - Where Optimization is Performed


System D Backtested Optimization Period with Real Time Data


System D Real Time Trade Data - After Optimization Period


Issuu converts static files into: digital portfolios, online yearbooks, online catalogs, digital photo albums and more. Sign up and create your flipbook.