Automated Rule Based

System Trading

V.S.

Buy and Hold

Index Funds

Kenneth W Sweet

 The securities markets are made up of millions of investors, traders and speculators making emotional decisions to buy and sell various securities at various times for various reasons. Because over time people and their emotions are the same, our research has found that if you use sufficiently rigorous methods to avoid hindsight, you can test a system and see how it would have done in the past and get a fairly good idea of how that system will perform in the future. Therefore building objective linear models of security and market behavior gives us our edge. 

 An interesting study was reported about in the book Decision Traps. This book  is about the process of decision-making. It puts forth the notion that objective decisions (i.e., rule based strategy trading) produce far superior results than other non-objective forms of decision making.

 In this book, nine different types of decisions were tested using each of the three different decision methods. The accuracy of the decisions was then compared and analyzed for effectiveness in predicting final outcomes. The investigator looked at different types of decisions, predicting grades, predicting recovery from cancer, performance of life insurance salesmen, as well as predicting changes in stock prices. He used three different decision making processes: an Intuitive Prediction Model, a Subjective Linear Model, and an Objective Linear Model. Interestingly enough, these can be compared to the 3 major categories of traders: discretionary, technical and strategy. 

 When making decisions based on an Objective Linear Model, a researcher will develop an objective model based on historical tests and observations to predict results. This is defining and using quantifiable data, running historical tests, and then using the results of the tests to predict future outcomes.

 For instance, the researcher would look at reams of physical data from leukemia patients, and correlate the data with how long the patient lived. After running the historical tests, the researcher would then obtain the physical data from a leukemia patient, and using the historical test data, attempt to predict how long that leukemia patient will live.

  This is exactly what a strategy trader using computer models for systematic rule based trading does. He runs historical tests and then uses that data to take a position in the market. He uses objective, quantifiable data tested historically to make his trading decisions. The table below shows the results of the nine different types of decisions that were tested using each of the three different decision methods.

of

Judgments

Intuitive

 In every case, the Subjective Linear Model outperformed the Intuitive Prediction Model but only by a small margin. If you look at predicting the changes in stock prices, the Subjective Linear Model only slightly outperformed the Intuitive Prediction Model. This correlates very closely with real world trading. Technical traders do only slightly better than discretionary traders and neither of them make much money. While the difference in expertise and experience between a discretionary trader and a technical trader is substantial, the resulting profitability is hardly noticeable. The real insight from this study comes when we look at the results of the Objective Linear Model. In every case, the Objective Linear Model outperformed both the Intuitive Prediction Model and the Subjective Linear Model. In some cases, the improvement was minor, and in others it was substantial. It is interesting to observe that the greatest improvement came when using the Objective Linear Model in predicting the changes in stock prices. Here  is a definitive study showing the benefits of objective decision-making as opposed to other forms of decision-making.

 Successful investment is really a matter of odds, and if you can compute the odds, you can find and test methods that could beat the market. Because we know that we don’t know. No matter what information you have, no matter what you are doing, you can be wrong. Therefore a good system will incorporate the following rules. First, never bet your lifestyle, from a trading standpoint. Second, if you know what the worst possible outcome is, it gives you tremendous freedom. The truth is that, while you can’t quantify reward, you can quantify risk.

 Instead of trying to control the market, the strategy trader lets the market tell him what to do. He Lets the market and his strategy take him long rather than his personally trying to predict or decide when to go long. A trader using a systematic rule based model realizes that he can’t understand the market, and that he can’t predict when the market will move, therefore he trades using a strategy in an objective and detached state of mind where he lets the system take him where it will when it wants.

  I find that evaluating systems solely on a calendar year basis is very arbitrary. What we really want to know are the odds for profitable performance in a holding period of any length. In the information that follows I will go over the concepts and basic steps that I use to design and build systematic rule based trading systems. Next I will show a few examples of actual systems I have developed and their results over various holding periods.

1. Business Cycle

 In order to understand macro stock movements, the reader must first be aware of business cycles. There are generally four stages to a business cycle: expansion, prosperity, contraction, and recession. The stage in which an economy is operating will have a significant impact on a company’s profitability.

 Beginning with a low point (recession), business activity increases until a period of prosperity is reached. Eventually the economy becomes over heated (inflation) and business activity begins to decline until a low point is reached again.

 The stock market acts as a leading indicator as smart money prepares for the next stage of the business cycle. Therefore, the stock market also goes through cycles, these cycles tend to lead the economic cycle.

 Just as the tide coming in will lift all the boats on the water, a rising market will tend to push the prices of all stocks higher. Company stocks can and do act independently of what the overall market is doing, however a great deal of stock performance is a direct reflection of the condition of the over all market.

3. Time Frame

The time frame in which you view your investments will also have an impact on your overall profitability, comfort level and expectations.

Below are two charts, both of the same security in different time frames.

As you can see, the shorter the time frame, the more the noise in the security movement and the harder it is to determine the true direction of price movement.

 Key ingredients of a successful trading system include the following:

1.     Entry signal or “ When do I buy?” (The smallest factor in a successful system.)

2.     Exit signal or “ When do I sell?” (Accounts for approximately 30% of overall profit.)

3.     Risk or position sizing, capital preservation, and capital at risk (The largest factor in a successful system.)  

4.     The profit expectancy of the system (If we know over a sample of 30 trades, the system will make an average profit of 7% per trade, then we can determine there is positive expectancy of $7 for every $100 in our portfolio.) 

Only trade systems with positive expectancies.

6. Scoring and Ranking potential buy candidates

Now we have a group of stocks to choose from. We have developed a system that has a positive expectancy with good risk management.

When the position sizing equation tells us to buy  five new securities and we have buy signals for 30 securities, buy the five securities with the highest score. This ensures always buying the best possible candidates at any given time.

A1 System

The chart bellow shows the equity growth of the A1 system in light blue. The dark blue line going through the middle of the chart represents the growth of the same funds placed in an SP500 fund. The red area shows equity drawdown from most recent peak.

9. System Scalability

Money managers realize that the effectiveness of their trading systems drop off rapidly after a certain level of assets is reached. Therefore, to demonstrate the scalability of the systems, I tested the A1 system as above applied to a $1 billion dollar portfolio (1/1/2003 –10/4/2004.)

Limiting “Buys,” to 5% of traded volume per week and priced at an average price for “Buys,” and “Sells.”

Relative Performance%  A1 System Equity, DJ-30, COMPQX, SP-500.

A1 System vs. Buy and Hold S&P 500. $1,000,000,000.00 Investment.

Profit distribution

Max. Adverse Excursion distribution

Max. Favorable Excursion distribution

Sweet Exhaustion System

I wrote a book about the system bellow (Light Gains… Greater Profits) The system applies a very short term trading strategy with an average holding time of approximately 5.6 days. In the test period of 1/1/2004-9/23/2004 this system achieved an annual return of 30.47% and a total return of 21.13%. The average profit was 0.59%, with 47.55% of trades profitable. There were 68 winning trades and 75 losers. The maxim system drawdown was –10.34%. The chart below shows an example of a buy (Green Arrow) on ESPD and then a profit exit (Red Arrow).

Sweet Exhaustion system 1/1/2004 – 9/23/2004.