August 2000 Newsletter

I. Financial Users – We Need Your Help!

If you subscribe to Technical Analysis of Stocks and Commodities Magazine, you have probably received your Readers’ Choice Ballot with the August issue. Remember to vote for the NeuroShell Trader in both the “Standalone Analytical Software>$500” and “Software – A.I.” categories. (You don’t even have to own it to vote for it!)

We appreciate your support! The higher our ratings here, the more sales we make, which means more resources devoted to helping you! The deadline is Sept 11, so please vote right away.


II. Writing Your Own Indicators in Power Basic for the NeuroShell Trader Professional and DayTrader Professional

Building Dynamic Link Libraries (DLLs) for your own custom indicators is possible with the NeuroShell Trader Professional or NeuroShell DayTrader Professional. You can use a programming language to build a DLL when the indicator gets too complex for the Indicator Wizard. There are two compilers we have used to build DLLs: Microsoft C and the Power Basic DLL compiler (Visual Basic won’t do).

Building DLLs in C is more complex than building a Power Basic DLL if for no other reason than that Power Basic is a simpler compiler to use. The learning curve for Power Basic is much shorter. On top of that, unless you are a pretty experienced programmer, Basic is a much easier language which still provides all the functionality we can imagine you may need to build indicators for the Trader.

We therefore believe that even beginning programmers can successfully use Power Basic and create their own simple DLLs in less than an hour.

To get you started, we have built three simple indicators in one DLL. You can look at the source code in a text editor to see how easy it is. Log onto with your Trader serial number and look in the New and Updated Examples section to download the examples with source code.

If you decide to build DLLs with Power Basic, either for your own use or to sell as a NeuroShell Trader add-on vendor, you can purchase the Power Basic DLL Compiler at for a reasonable cost.


III. Comparing Neural Nets with Regression Analysis and Other Out-of-Sample Considerations

The following tip has recently been modified on our advanced users and tech support web site (

When you compare nets with regression (and other modeling techniques for that matter) you have to be careful to compare apples to apples.

In all of our documentation, we emphasize evaluating your neural nets on “out-of-sample” data, not the data with which you train your net.

However, practitioners of regression analysis often do NOT do this. They report results of the regression “training set” and sometimes fail to use out-of-sample testing. This is as misleading as reporting the results of the neural network training set.

Therefore, when you compare nets with regression, either make both out of sample or make both in sample. Use exactly the same data, or the comparison isn’t fair.

NeuroShell 2 users have to be careful about another thing that NeuroShell Predictor and NeuroShell Classifier users don’t have to worry about: calibration. Calibration in NeuroShell 2 means that a test set is extracted from the training set. This is never done with regression analysis and most other modeling techniques. So, to make sure you are comparing apples to apples, turn OFF calibration and use the whole training set for training, just as you do when using regression.

Here’s another thing NeuroShell 2 users have to worry about. Activation functions in the output layer can make quite a difference. When predicting numeric amounts, the linear output activation function is usually the best to use in the output layer (you’ll probably need low learning rate and momentum). The logistic output function is best for classification problems (e.g., comparing to logistic regression analysis).

Note on the genetic method of the NeuroShell Predictor and Classifier: this unique method always trains everything in an out-of-sample mode; it is essentially doing a “one-hold-out” technique, also called “jackknife” or “cross validation”. If you train using this method, you are essentially looking at the training set out of sample. The same is true if you turn on enhanced generalization when you apply the net to the training set. This method is therefore great when you do not have many patterns on which to train. However, the training set error statistics will not look as great as some method where one hold out is not being performed.

While we’re on statistics, we’ve seen many cases where “R squared” in our products is compared to “r squared” in other products, especially regression. They are different measures with different formulas. The tricky thing is that they result in the same value when using regression! In non-linear models such as neural nets, they AREN’T the same thing. Be very careful here!

Note to NeuroShell Trader users: although the NeuroShell Trader Professional can fire NeuroShell Predictor and Classifier nets, the “one hold out” method will not help you. That is because in financial predictions it is not really out of sample to predict day X when day X-1 and day X+1 are in the training set. You can’t trade that way. That is why the Trader does not allow any “random extractions” of evaluation data.

Note to NeuroShell 2 financial users: since NeuroShell 2 does allow random extractions, your predictions look much better than they really are if you have a randomly extracted production set! Don’t use random extractions for either the test set or the production set.

Scaling could cause you to be comparing apples to oranges. The Predictor, Classifier, and Trader all scale data before building a model; there is no way to turn it off. So your regression models should be scaled too, with something like the Z-score (that is standard statistical practice). You can turn off scaling with NeuroShell 2, but most of the time regression models are scaled, so use it. But NeuroShell 2 also has clipping, which should be off, because regression models usually don’t clip. To make NeuroShell 2 use Z-score, use the scaling function = mean + or – 1 standard deviation and turn clipping off.

Now here’s the last thing of which to be wary. If your data is essentially linear, there isn’t anything that can beat linear regression. You shouldn’t even be using neural nets for linear data and no comparisons can ever be fair. If you are supposed to be comparing methods, then make sure you include non-linear data. Neural nets are highly non-linear models that will excel with non-linear data. If you have sparse linear data, neural nets will try to fit all the noise and make a linear model out of it, and then they’ll surely look worse than regression. If you suspect your data might be linear, then make sure your nets are linear too, especially if there isn’t a lot of training data (less than 300-500 training patterns) or if there are more than 5 inputs. The NeuroShell Predictor and Classifier and Trader Professional are made linear with zero hidden neurons. There is no way to force the genetic method to be linear. In NeuroShell 2, all activation and scaling functions have to be linear for a linear model.

Final note to NeuroShell 2 users: if you have read all of the above and you feel overwhelmed by issues like test sets, activation functions, scaling functions, clipping, learning rates, momentum, setting hidden neurons, etc., then you now know why we have been trying to get all of our NeuroShell 2 users, except college professors, to start using the NeuroShell Predictor and Classifier or the NeuroShell Trader. The latter programs give you better models (when apples are compared to apples) without all the tweaking and knowledge required. (College professors have to teach the classic algorithms everyone else uses, and tweaking gives them more to teach anyway!)


IV. New WSG Discussion Forum

Many of you have been asking for a way to discuss issues with other Ward Systems Group (WSG) customers. We are happy to announce the opening of the WSG Discussion Forum on It will open on or about Tuesday, August 22. It will be a moderated forum where users can discuss their applications, seek help, or offer solutions specific to using WSG products.

The Discussion Forum is not meant as a substitute for technical support, which is still available by sending email to or by calling (301) 662-7950. We may or may not respond to postings ourselves, as time permits.

To enter the WSG Discussion Forum, simply go to and log in with your product serial number. On the next screen that is displayed, click on Discussion Forum. The Discussion Forum is divided into different discussion threads organized by product. You can view commentary on any WSG product, but you can only post to the product corresponding to the login serial number.

This is a forum moderated by WSG. It may take a day or so for postings to become available, and WSG reserves the right not to post everything that is sent. The forum is created in the spirit of mutual education for all.

Users desiring to post to the forum must supply their real name, forum name, and email address. The user’s forum name may be different from the user’s real name. Only the user’s forum name will appear on the post. Real name, email address, and product serial number will not appear.


V. Beta Testers Needed for Satellite and Other Data Feeds

We are looking for some users to beta test a version of the NeuroShell DayTrader that will interface with the Universal Market Data Server (UMDS). In order to beta test, you must have the NeuroShell DayTrader Professional, the UMDS software installed, and have one of the feeds listed on this web page: PLEASE NOTE: This will be a beta test using version 2.0 of UMDS which does not support Internet feeds at this time.

This is a beta test only and the features tested may or may not be implemented.

If you are interested in participating in this beta test, please send us an email at , with your name, address, phone number, and DayTrader serial number, as well as the name of your current real time data feed. We will contact you with details on installing the necessary files once they become available.


Was this article helpful?

Related Articles