Stocks and Commodities: Mar 2009 - Trade System Evaluation
Posted: Wed Feb 18, 2009 10:54 amThis post is for those interested in exploring "Trade System Evaluation", an article by Donald W. Pendergast Jr. in the March 2009 issue of Stocks and Commodities magazine.
You can import the necessary Setups by right-clicking on the following file and selecting "Save Target As...":
After the above file has been saved on your computer, use the File > Import Database menu to locate and import it. You will then have access to:
2 Strategies named:
MACD Reversed Crossover System
MACD Standard Crossover System
There are two primary concepts described in this month's article, both of which have been available to StrataSearch users for quite some time. The first concept is that indicators often work better when they are reversed than when they are used in a traditional approach. For his trading system, the author discusses the MACD. In the traditional approach, the trader will buy when the MACD rises above its signal line, and sell when the MACD drops below its signal line. The author suggests that the reverse approach is often better: buying when the MACD drops below its signal line, and selling when the MACD rises above its signal line.
The two strategies provided in this month's plug-in contain these 2 MACD approaches. Users can experiment with these to see if the author's theory holds true. Automated searches in StrataSearch already contain both the standard and the reversed approach, both of which were implemented with a range of variables. The trading rule "MACD: 1" uses the standard approach, while the trading rule "MACD: 2" uses the reversed approach.
The second concept described by the author is the use of a Monte Carlo simulation in trading system evaluation. In fact, he uses a Monte Carlo simulation to evaluate the performance of the standard and reversed MACD systems, saying that the Monte Carlo results provide a better prediction of how the trading system might perform in the future than other performance statistics. StrataSearch allows you to evaluate the Monte Carlo results on the Monte Carlo tab of the Detailed Analysis. StrataSearch users can also include filtering or scoring based on Monte Carlo results in their automated searches. For more information, including examples, open the StrataSearch Help, Advanced Features chapter, and open the page on Curve-Fit Protection. There you'll see examples for Monte Carlo Average Annual Return and Monte Carlo Average Drawdown.
While this month's article may not provide anything particularly new or unique to StrataSearch users, it does provide confirmation of a couple of techniques we've been using for years. In particular, the reversed implementation of an indicator is often more effective than the traditional implementation, and the Monte Carlo simulation can be a helpful statistic when evaluating trading system performance.
Monte Carlo Chart of the MACD Reversed Crossover System:
You can import the necessary Setups by right-clicking on the following file and selecting "Save Target As...":
tasc_mar2009.sse- (380 Bytes)
After the above file has been saved on your computer, use the File > Import Database menu to locate and import it. You will then have access to:
2 Strategies named:
MACD Reversed Crossover System
MACD Standard Crossover System
There are two primary concepts described in this month's article, both of which have been available to StrataSearch users for quite some time. The first concept is that indicators often work better when they are reversed than when they are used in a traditional approach. For his trading system, the author discusses the MACD. In the traditional approach, the trader will buy when the MACD rises above its signal line, and sell when the MACD drops below its signal line. The author suggests that the reverse approach is often better: buying when the MACD drops below its signal line, and selling when the MACD rises above its signal line.
The two strategies provided in this month's plug-in contain these 2 MACD approaches. Users can experiment with these to see if the author's theory holds true. Automated searches in StrataSearch already contain both the standard and the reversed approach, both of which were implemented with a range of variables. The trading rule "MACD: 1" uses the standard approach, while the trading rule "MACD: 2" uses the reversed approach.
The second concept described by the author is the use of a Monte Carlo simulation in trading system evaluation. In fact, he uses a Monte Carlo simulation to evaluate the performance of the standard and reversed MACD systems, saying that the Monte Carlo results provide a better prediction of how the trading system might perform in the future than other performance statistics. StrataSearch allows you to evaluate the Monte Carlo results on the Monte Carlo tab of the Detailed Analysis. StrataSearch users can also include filtering or scoring based on Monte Carlo results in their automated searches. For more information, including examples, open the StrataSearch Help, Advanced Features chapter, and open the page on Curve-Fit Protection. There you'll see examples for Monte Carlo Average Annual Return and Monte Carlo Average Drawdown.
While this month's article may not provide anything particularly new or unique to StrataSearch users, it does provide confirmation of a couple of techniques we've been using for years. In particular, the reversed implementation of an indicator is often more effective than the traditional implementation, and the Monte Carlo simulation can be a helpful statistic when evaluating trading system performance.
Monte Carlo Chart of the MACD Reversed Crossover System:
- tasc_mar2009.jpg (118.03 KiB)