Amibroker is great and cheap
March 16, 2006 by Trader Rich
As I’ve written in the past couple of weeks, I’ve been doing a lot of backtesting. I’ve used esignal (my data provider), metastock, and now amibroker to automate the backtesting process. If you ask any experienced and successful broker, they will say that you are absolutely crazy to not back test and forward test your system before going live with it. Here’s a quote from an experienced trader, chaffecomb:
I’ve absolutely no idea how some people can trade without first backtesting, however that’s just the way I’m wired! …
Check out his website. He tracks multiple trading systems that he developed and basically trades on autopilot. http://users.bigpond.com/morleym/index.htm
Out of the 3 systems I’ve used for backtesting, I really think that amibroker is the best and it’s the cheapest. Here are the pros and cons I’ve noticed for each:
Esignal
Pros: Nice reports and graphs post-backtest; the ability to backtest already written custom indicators with minor modifications
Cons: Expensive; no ability for optimization
Metastock
Pros: Full featured suite of already made trading systems
Cons: I found the entire product very confusing and I’m a programmer; Expensive
Amibroker
Pros: Intuitive user interface; fast; ability to optimize backtesting
Cons: they don’t provide data so you need to download free data or use your currency provider (esignal plugin)
The most important backtesting functionality is the ability to optimize your trading systems. For example, what if you wanted to backtest a moving average crossover. With esignal, you would have to manually change the moving average periods you want to test which would require an individual backtest for each moving average pair. With amibroker or metastock, you can use variables and have the program go through an entire range of moving averages all in one shot. It really is a powerful feature.
I’m using amibroker now and I don’t see how I would ever use anything else. In addition the full version was only about $250.
Popularity: 4%
Loading ...
Markets are dynamic not static.
Hi Rich. I posed three questions to traders this afternoon. Thought you might be interested in reading responses:
:(Questions to myself: Why would anyone regularly trade anything but the four majors?
Ans: Good question! Because another pair might have a better data structure at the time. Or, because it might have better Magnitude stability. Most majors don’t fit this bill, only a handful are stable enough for some traders to trade inside their systems. Not all systems are the same and therefore, not all market behavior will be detected and measured in the same way by all systems. –NaysayerTrader
:(Question: Markets are dynamic not static. Has there ever been a mechanical signal “system” that works!! (90% win rate = Holy Grail)?
Ans: Good question! Mine is currently running at 100% accuracy over the past 47 trades (this week’s trades have been posted on the forum) and 94.82% accuracy over the past 280 trades with the current Day Trade in tow. It really all depends on the level and quality of your anaylytical research and the degree of creativity one brings to the process of system design, architecture and component integration. If done correctly, one can produce a very accurate Grail form which to profit from consistently over many years with the right degree of balanced tweaking to meet new market demands. In addition, if the system was “designed” from the outset to measure a “dynamic” market, then by definition its output will reflect that and be in line with a dynamic market.
If the system was designed to measure “static” markets (which I know of no system that meets that qualification), then said system will reflect such market conditions. It really all comes down to reasearch and correctly applying concepts discovered. Lastly, some systems are based on conventional 30 year old TA, while others are completely new approaches that use no traditional TA — in fact, some systems re-write the rules when it comes to TA development, so unless one has been exposed to “all” systems, then it might prove rather difficult to lump them into the same “Grail” classification. Trader
:(Question: Why would a successful trader use someone else’s system to execute trades (unless question 2 above)?
Very good question! No one single trader knows it all. Thus, any single trader is capable or learning something new from any other trader. There might be an instance when trade “A” is just one indicator away from raising the level of his system’s overall accuracy to a point that separates himself from the rest of the crowd, or at least places his trading ability into a new class of success. That one single idea can come from anywhere, and be sparked from anything.
Example: I was once on a flight (and I have many stories over the years just like this one) from Toronto to SFO coming back from a business trip. While on the plane somewhere over mid-America, I looked out the window and picked up the horizon. As I’m looking at the horizon thinking what a wonderful day it was to be coming back home after three long weeks, I was hit by a lightening bolt.
I was contemplating in my minds-eye the curvature of the earth at the horizon and envisioned that curvature [b] increasing [/b] with altitude and then [b] decreasing [/b] with a loss of altitude. That one single thought lead to the development of an entirely new system module that I call the [b]EventHorizon. [/b] It later became (with much tweaking over the years) one of the most stable and most accurate trading signals in my entire arsenal. One never knows where the next big idea will come from.
Having said that, I would also say that all traders should learn how to trade effectively on their own and develop their own “system” for trading whatever that system might be. Some people are gifted at being able to play the news on the Fundamental side of the business while others are gifted with working with raw numbers and raw data. There is no one single way to become a successful trader, but using someone else’s work product to trade is not necessarily a negative–if that perosn is pushing the old system to new heights of achievement. Otherwise, I’m in the camp that says, Technical Traders should know their system inside and out before trading it. Trader
Greg. Great stuff as usual. I appreciate the post once again….