Once again I'm trying to squeeze something useful from Adaptrade Builder. Let me remind you, that this program itself creates trading systems for any chart, or a portfolio of charts. Those interested can take a fully functional trial for 14 days. Since I am after installing Windows 8 haven't taken trial yet, then yesterday I decided to try my luck again.
The principle of the program is simple — data is imported in text format, indicators are selected, which will be sorted out by the program to create a system, signals are selected (stop, take profit, trail, market, etc.), who will also participate in the creation of the system, the desired parameters are selected, drawdown type, profit factor, etc., the values of which you want to see in the system and many more different settings. Then, when everyone set up, press the start button and the program begins to build a system using biological methods of the population, mutations, etc.. As a result, you get a lot of systems, which can be sorted by different parameters. I prefer to choose by the visual method according to the evenness of the upward equity.
Also, the program generates, optional, code for TradeStation, Multichart, Omega 2000i or Metatrader 4. Thanks to this code, you can see the logic of the system and understand it.. As I understand, all systems are built according to a template:
Filter for inputs, type & quot; Indicator1 > Indicator2" etc.
Entry level, of type & quot; Indicator Level + Number * ATR
Outlet filter
Exit level, feet, trails, take profits
Sometimes the entrances / exits are just around the market. You can also apply either symmetric logic for longs / shorts, or asymmetrical.
Then, the generated system code can be inserted into one of the listed programs and check how the system actually works, see deals on the chart.
All in all, after two days of intensive use of the program, came to the conclusion that such a uniform structure of the generated systems limits the potential of the program. Actually, it just adjusts the results using parameters, and not some logic of using indicators. For trending instruments, the level is calculated for the breakout by a stop order, and for instruments, tend to fall back to the mean, the level is calculated on a rebound with a limit order. But this is understandable even without a program and only in order to understand how to trade a specific instrument, there is no point in using the program.
Tried to create a system for a futures portfolio, but the program turned out to be powerless — no standing system was created even by fitting. Certainly, it's clear, which is much more difficult to fit for a portfolio than for a single instrument. For a single tool, almost any system can be adjusted simply by simply optimizing the parameters.. But with a briefcase, such a number does not work.
Although, it cannot be said that the program is completely useless. With a careful approach to studying the logic of the generated systems, one can draw some conclusions about the tendencies and nature of the tested instrument., and, even new thoughts and ideas appear about new systems. So that, may be, the program as a tool for generating ready-made systems and does not roll, but it can serve as a tool for studying the inefficiencies of a particular futures, index, etc.
