Date: October 26, 2005 14:00GMT 10EST
Expert: David Lyder, Co-founder of iExpertAdvisor, LLC.
Topics:
- Why everyone trader should have one?
- Various types of systems
- Strategy for building a system
Who is David Lyder?
David Lyder is a co-founder of iExpertAdvisor, LLC, as well as the author of Automatic Alpha: How to Build a Winning FOREX Trading System. iExpertAdvisor is a newly formed company dedicated to providng information and ideas about FOREX trading systems. Visit us at http://www.iExpertAdvisor.com .
Speech Material:
If someone offered you $10 a day, everyday, what would you say? Would you reply: No, thanks - $10 isn’t that much.
If your smart, you’ll take the money. Why not? After all, $10 is $10, and it adds up quickly. Running a profitable mechanical trading system is similar to someone giving you money almost everyday. Whether it is $10, $100, or even a $1000 – the work required after building the system is the same. Simply turn on your computer and start the program.If gaining access to an automated trading platform was expensive, that might be a legitimate reason for a trader to not have a mechanical system. After all, who wants to accept more risk. Now, in addition to beating the market, you also need to generate enough profit to cover the cost of the trading system.But a free platform is available. And as a FOREX trader, you are already aware of demo accounts and mini accounts.
So here is the situation:
You have a free platform capable of developing, testing and executing an automated trading system.
You have a demo account so you can run the system live and evaluate its performance before risking any money.
When you are satisfied with the system’s performance on the demo account, you can open a mini account and risk a limited amount of capital.
Given the upside of running a mechanical trading system, there is no reason every FOREX trader should not run at least one mechanical system.
Whether you are a technical trader or a fundamental trader or a discretionary trader – why not have a mechanical trading system running in the background continuously generating profits for you?
And here is the bonus: the skills used to develop a mechanical trading system will enhance your skills as an overall FOREX trader. Once you have learned the framework used to develop a robust mechanical trading system you will naturally apply the same methods to all of your trading decisions. This will make you a better FOREX trader. Okay, so you realize and appreciate the benefits of building and running a mechanical trading system. Well then, how does a trader go about building a mechanical trading system? That is what this Q&A session is about.
Various Types of Systems. There may be as many kinds of mechanical trading systems as there are traders. To name a few types of systems:
Breakout System
This system determines a breakout price level. If the price of the currency pair breaks through this level, the system trades in the direction of the breakout.
Reversal System
This system looks for signs that a price movement is near exhaustion and will soon move in the opposite direction. When the price exhaustion is confirmed, this system trades in the direction of the reversal.
Indicator System
An indicator system simply follows the rules of the particular indicator. Wells Wilder’s oscillator systems are often built as mechanical trading systems.
Trend Following System
A trend following system determines the existence and direction of a trend. If a trend exists, the system trades in the direction of the trend. Any of these systems can prove profitable. But be warned, the road is littered with new traders who were fascinated with the “holy grail” system. I can not even count how many traders I have worked with that were
- just looking for a scalping system that
- just went into the market a few times a day and
- just scalped about 20 pips everyday
- A system like this can be difficult to build. (Especially if it is your first system). There are a few reasons:
- with so many trades, trading costs become a significant factor;
- while price movement in general may not be random, the lower the timeframe, the more random it behaves;
- these systems are difficult to backtest – most historical tests do not use tick data, which is required for testing this kind of system;
- many of these systems usually do not work for long.
Actually, there is evidence that suggests the only type of trading system that proves to be profitable over the long run is the trend-following system.
Gary Hirst, of Hirst Investment Management Inc., conducted research on technical analysis (TA) in 1991 and found that most TA systems, besides trend-following systems, eventually give back their profits. (See this article).
This does not mean that many of these TA systems are never profitable. In fact, many of these systems have a good run – maybe 5-10 years of profitability. Then they stop working.
Therein lies the problem with building a TA system as your first mechanical system. You don’t know when the system will stop working. (Probably when you start trading it.)
So there are at least two good reasons to build a trend-following system (especially if it is your first mechanical system).
Trend-following systems have produced steady and consistent results historically. The market has been shown to distribute returns unequally – that is with long fat tails.
A trend-following system is easy to divide into separate parts; each part can be developed and tested separately.
Strategy for Building a System
A good strategy for building a mechanical trading system is to break the system into parts. Each part is then individually developed and tested. A trend-following system is easily separated into the entry and the exit.
The Entry
The job of an entry in a trend-following system is to get the trade started in the right direction.
An effective method for testing the entry is to simply exit the trade after a fixed number of bars (the exit-bar test) and record the winning percentage. The trading costs are not considering in the test. Neither is the profitability. The only metric considered is the winning percentage:
Winning percent = (number of winning trades)/(total number of trades)
The winning percent must be greater than 50% - otherwise the entry in no better than random. Generally, the higher the winning percent, the better the entry.
After an acceptable entry has been proven through the exit-bar test, the entry can be tested for robustness.
In this context, a robust system is defined as a system whose input can be varied without significantly changing the output.
Generally speaking, a system with fewer input parameters is usually more robust than a system with many input parameters.
If an input parameter can be removed from the entry, without reducing its effectiveness, the parameter should most certainly be removed. This will increase the robustness of the entry.
The entry can be tested for robustness by varying the input parameters greatly, and retesting the entry using the exit-bar test.
For instance, if the entry uses a 10 period moving average (MA) as an input parameter, test the entry using a 5-period MA and then a 20-period MA. If the winning percent is still acceptable, the entry is probably robust
The Exit
Most traders place too much emphasize on the trade entry. Again, the sole purpose of the entry is to get the trade started off in the right direction. It is the responsibility of the trade exit to actually make a profit.
The quality of an exit is judged by its profitability – so when performing exit testing all trading costs should be included in the test.
The simplest exit strategy is the fixed stoploss and fixed take profit. This is often used as a benchmark against other strategies.
An exit strategy can be based on a technical indicator, a time factor, and of course a price-level.
The most popular price-level based exit strategy used with trend-following systems is the trailing stop.
There are many creative ways to implement a trailing stop strategy. The simplest is for the stop to trail the price by a fixed number of pips. More advanced trailing stop strategies may consider the highest and lowest price values in the last few periods as well as the time the trade has been open.
In order to test an exit objectively, the entry must remain constant. With a constant entry, each exit strategy can be tested and the results can be compared and evaluated.
In general, fixed stops usually provide a greater profit than trailing stops. However, trailing stops usually have a higher winning percentage. The trading system developer must choose what metric is more important: profits or winning percentage.
Live Perspective Broadcast
Date: October 31, 2005 15:00GMT 10EST
Expert: Derek Frey, Head Trader of Odom & Frey Futures & Options
Topics:
Market reaction to news!
- This week's data and events
- Key support and resistance for major markets
- Derek's insight on the week inreview and the week to come
Currencies
- Euro
- Dollar
- Pound Sterling
Who is Derek Frey?
Derek Frey has been a futures trader since 1989 and is Head Trader and partner at Odom & Frey Futures & Options, a firm that specializes in high probability, defined risk option spread trades for the futures markets. Derek is also co-author of O&F News & Views, a weekly newsletter that is published by several publications and over 20 financial websites including FXStreet.com.
Speech Material:
Market Reaction to News
As I suspected last week the Stock market did not follow through to the downside and is now trying hard to rally back up above 10,400. We are recommending a short term bull call spread at this time by buying the Nov. Dow 105 call and selling the Nov. Dow 107 call as a spread. There is a very strong seasonal tendency for stocks to rally for about the next four weeks. The trade I suggested above is a great way to be positioned to take advantage of this trend if it develops while keeping risk in check.
Key Support and Resistance for Major Markets
Bonds: Bonds Managed to break down this week after a brief dead cat bounce late last week. I continue to be concerned by the excessive small spec short interest. We are doing our FOMC Bond strangle again for Tuesday’s FOMC meeting. I expect a rather dramatic increase in volatility in Bonds after this Fed. meeting due to all the chatter about this being the “8th inning” again. Derek Frey 10-28-05
Insight on the Week in Review and the Week to Come
With a two year low reported in U.S. consumer confidence along with a 2.1 percent plunge in durable goods orders, eyes will be focused on Friday’s Gross Domestic Product figures. Nascent doubts remain on the strength of the world’s largest economy and the validity of rising interest rates.
Currencies/Economic Indicators
The Us Dollar did fail to follow through to the upside this week. This should have given metals boost but was not enough to get them to breakout to new highs. Gold continues to consolidate around the 475 level, while silver is doing the same around the 7.80 level. Copper too is consolidating around the 180 level. Palladium has been the exception as it continues to make new highs. Platinum is trying to follow Palladiums lead but so far has struggled to break out. All of this points to the strong possibility of a continuation of the overall bull trend we are seeing across all metals. Since we have seen such strong consolidations lately it is reasonable to expect that we are only days away from a break out. This Friday’s late dollar rally is likely to fail next week just like last Friday’s dollar rally failed to follow through this week Derek Frey 10-28-05
EUR/USD: The Euro did try to stage a rally this week but was largely held back by the rather stagnant Dollar. I do expect further downside on the Dollar so the Euro should rally this coming week. Wait for a move above 122.00 to get long
USD/CHF: With the Dollar failing to follow through the Swissy also failed to breakout above 130. I do not expect much strength in the dollar this coming week so look for this market to continue to fall.
GBP/USD: The Pound did begin moving towards 180 but fell slightly short. I do see 180 being broken through this coming week.
USD/JPY: I continue to be bullish this market and expect to see a move towards 117 this coming week
AUD/USD: My stop and reverse order at 74.25 so far has held and I do think we should see the beginning of a rally this coming week
USD/CAD: The Canadian broke out mid week only to pull back 300 points in the next four days. I look at this as the final shakeout before a dramatic rally that caries us above 120.00 near term
USD/MXN: The Peso continues to drift sideways this past week. I am bullish this market but not overly bullish. I would prefer the Canadian if I could only be long one currency
Continue reading "Live Perspective Broadcast" »
Posted by ClaireFlaherty in Expert Q&A Sessions, Market Viewpoint & Live Commentary | Permalink | Comments (0)