Trend Following Systems

So I’ve started playing with Neuroshell, while simultaneously attempting to read a few trading books to get some ideas. I also picked up a copy of Market Wizards by Jack Schwager to get some inspiration – trading research can be extremely frustrating.

I spent yesterday trying out the obvious: conventional trend-following systems such as moving average crossovers and breakouts – stuff that the Turtles and other trend-followers worldwide use. The results were expected: while the systems performed admirably in in-sample data, they failed miserably in out-of-sample data. Could trend following really be dead? The markets spend more time in sideways ranges than in trends, and those sideways markets just kill the equity curve.

I’ll try some new oscillators tonight such as MACD and Stochastics to see if they’ll work.

I’ve also come across a book: The ART of Trading by Bennett A. McDowell. McDowell has a very simple strategy that is based on the breakout system that so many people follow. The only difference is that it recognizes what it calls “Pyramid Trading Points” – triangles that point to the direction of the trend. These triangles are generated during trend pullbacks. The main advantage of them is that they offer clear entry and stop-loss points, which let you calculate your position size in advance. (See picture)Pyramid Trading Point

The tough part is programming it in Neuroshell – I’m having trouble defining the triangle apex as a reference point so that it would trigger an entry when the price exceeds the apex. It seems like a relatively simple idea, and it makes sense since many traders trade like this – by identifying the key support and resistance points. Trying to figure out how to program it seems like a worthwhile way to spend my time, but it’s annoying because of the way Neuroshell is structured – having to search and click parameters seems a lot more confusing than typing it out. I’ve asked my dad for help.

As for now, I’d like to focus on the other trend-following systems – testing them using both ProTA and Neuroshell to make sure that I did not miss anything. If I can’t find a way to make them work, I’m going to have to move forward to find something else. Such a pity, since trend-following is something that I’ve understood the best for the past year.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s