Custom Search

The Final Step

Now that we`ve demonstrated the capabilities of a parallel function, the next step is to develop an automated system to take advantage of these functions.

Our goal is to create a system which will use parallel functions to constantly adjust its trading patterns in response to the ever changing personality of the market. In a sense, these systems could be thought of as self optimizing.

The programming here is based around the use of loops and arrays to test all the available options of inputs as selected by the user. It is also possible to restrict the self testing by the system to a set parameter, such as the number of days, number of trades, etc. For instance one could self optimize the system for the last 30 days, the last 10 trades, the last 5 winning trades, etc. The possibilities for self testing are limited only by ones imagination once the parallel function is created.

In all cases, it is possible to program the system to test itself against a pre defined set of back data and, after determining the optimal set of variables for the current period, install these variable sets into the system equations so that they will be used for the next trade.

The advent of TradeStation 2000i has broadened the scope of this technique exponentially with the ability to manipulate arrays from within a user function and the elimination of the 64 K file size for system code. As time goes on we will no doubt find expanded uses for parallel functions with the vastly increased capability offered by 2000i.

Summary

Parallel functions can be a useful tool for both system development and testing.

The ultimate use for this tool is found in the creation of systems which are ultimately self adaptive in that they can use their own reaction to current market conditions to change their trading parameters on the fly.

It cannot be emphasized enough that this technique is by no means a fix all for a mediocre or poor system. In fact, it will probably decrease the performance of such a system.

In the final analysis, the user will find this technique useful as an additional tool to make his or her trading more reactive to current market conditions.