Long Term Trend and Widening Stops

Old Filter
New Filter
Even till now I still am hindered by inexperience in systematic trend following. A few days back when reading a book, I came up with a new filter which I was convinced it would offer a better view of the trend. I implemented this new filter and it drastically improved results. The older system used a moving average cross over filter for identifying the long term trend while the new filter used a another form of moving average to do the same job. Above are the attached images of the differences.

Sharpe Ratio:  +34%
MAR Ratio: +48%
Average Win / Average Loss: +35%

The significant difference was due to a better and more efficient way of identifying trends as it greatly decreased whipsaws. Equity curve was a lot more smoother.

Enough with the filter stuff and coming back to the fundamentals of trend following. My first system used a 4 ATR stop from start to stop. I gave little attention to selecting the numbers but then the idea behind this was to ensure winning trades get their way instead of cutting profits short. Attached below is monthly return distribution.

New Filter System

As you can see, the distribution has a shape like a positive skew. There are outliers to the right and it is more than 2 times the left most bar.This conforms to the definition of trend following as winners are bigger than losers. How can I efficiently let my winners run further while cutting my losses shorter?

The idea here is to not focus on entries and the accuracy of your system. The main problem for me in the transition from discretionary trading to quantitative trading is the bad habit of wanting to make as much money as possible through being right as much as possible. Just when I wanted to tighten my initial stop so to reduce my initial risk giving room for more contracts for my big winners, it so happened that it greatly increased the drawdown. Why? Well its simple, tighten stops for trend following system usually means increase stop outs when market volatility increases unexpectedly. There must be sufficient room for winners to be able perform. Below is the result of widening both initial and trail stop.

As you can see, although the distribution lost all the double digit outliers on the right, the central mean shifted more to the right which means better returns coupled with reduced variation.

What I learned…
1. stop focusing so much on the entries
2. start focusing on letting winners run and cutting losers short
3. trade with the long term trend

good trading,
M

Advertisements

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s