May 17, 2018
IN THIS ISSUE
7 MODELING PITFALLS THAT CAN COST YOU MONEY
CHAT ABOUT SUPPLY AND DEMAND TRADING SYSTEM
7 MODELING PITFALLS THAT CAN COST YOU MONEY
Here are seven common modeling mistakes we have learned to avoid in our over twenty five years of assisting
customers with how to apply Artificial Intelligence techniques to building trading models.
1. Don’t insist on only modeling a particular security
Not all securities are easy to predict. If you are not able to predict your favorite issue after a while, let
it go. It’s a big market, and there are a lot of ways to make money.
2. Don’t expect to build one model that is correct 100% of the time
Markets are constantly changing. You will be successful if you can build models that make money most of the
time.
3. Don’t overfit your model to training data
In other words, don’t make a model that works well when it is being built with known answers, but which
becomes unprofitable with new “out of sample” data, or in trading. The biggest mistake made by users of both
neural nets and other modeling techniques is to use too many inputs. You’ll notice most of our example models
only use 5 inputs.
4. Don’t worry about neural net structure and settings when building a prediction model
The only fact you need to know is that nets make predictions about the future based on what happened in the
past when similar input patterns occurred. The Holy Grail: You must give neural networks input patterns
during training that will repeat in the future. Furthermore, when these patterns repeat in the future, the
price movement that follows (i.e., the output) should be like the price movements that followed these same
patterns when they occurred in the past.
5. Don’t pick issues that are driven primarily by fundamental variables if you are using technical variables
Enough said.
6. Don’t go back too far in time
If you try to bullet proof your model by loading too much historical data, the input patterns found during
training will be substantially different from those showing up today. Even if you have great inputs that are
normalized over time, it is likely that as markets change, so does the effectiveness of your input variables.
7. Don’t put all of your eggs in one basket
How you trade with your models may be as important as the models themselves. No model will be perfect in the
financial arena. It may even be the case that the issues it works well with will not be consistently the
same. Fortunately, the NeuroShell Trader has a feature wherein you only need to build your indicators and
predictions once to apply them to a whole series of your favorite issues. You just place these issues in
different pages of the same chart. Then on a given day, adopt a trading strategy like purchasing the top 10%
as predicted by the model, and selling the bottom 10%.
____________________________________________________________________________________________________________________
CHAT ABOUT SUPPLY AND DEMAND TRADING SYSTEM
Join the discussion at nstsupport.wardsystemsgroup.com
We have just posted an example chart on the forum for building a Supply and Demand Trading System with the
Turning Points support and resistance indicators. Check it out in the Trading Strategy Development &
Backtesting section of the forum. Click here to download the chart. You must have the Turning Points add-on to view this chart.
If you’re new to the forum, you have to create a user name and password. These credentials are different
from your Trader serial number and password.