Skip to content

What to look for, what not to: Some pointers on system development

July 5, 2009

Quite some time back, I posted an article here, with this title, “Developing a Trading System”. It discussed on the psychological implications of trading a system. It never asked about the profitability[as in hard $] of a system, but instead asked some hard questions about its drawdowns, losing streak and average MAE. That post is more of an end stage question rather than a first step method.

This time, I am concentrating on something which comes right in the first steps. How do you really go about  fiddling with an idea to the first steps of investigating it seriously? And what next?

Before, I begin, I would like to reiterate an analogy. There is a striking similarity between markets and ecosystem. The trading systems being hunters, chasing limited $$$ for survival. If that is so, which animal should a trading system represent? Elephant[Fundamental Investing, might be, not sure, slow,regal,patient and huge]? Jaguar?

The issue is, your trading system can  be any one of them, yet in the situation of meteoric catastrophe, most of them would be wiped out. Save one.
The Cockroach.

A cockroach can survive even a nuclear holocaust, its cretin shells robust enough to save it from most of the catastrophes, man made or natural. Their hardiness, can keep them alive without food or water upto a month, and can survive on limited resources like glue off a postage stamp, the sebum left in a fingerprint.

The final idea is this, you want your trading system to be as hardy as the cockroach, surviving and treading a wide range of markets and market situation well enough, so that any change, catastrophic or gradual doesn’t do any harm to its tradeability.

And this forms the crux of the post. You have got an idea, say,  a simple day trading trend following idea, like go long when the price shoots above the previous day high, with stop as that bar’s low and short when the price shoots below the previous day low and with stop as that bar’s high. Once you get the idea and you code it, the first temptation is to check if its profitable, how much it is, etc etc etc.
Well not yet. Don’t do that, now. Instead what is suggestible is run the vary same strategy over a wide range of price series. Not all will be profitable, right now. Some will be, some will be not. But that’s okay, its a good idea to keep the profitable to non profitable markets ratio somewhere between 40% to as high as 60%.

Yet, look for the risk adjusted returns in all the markets in the sample set. None should stick out, individually. What effectively we are looking for is not a specialised animal but a ever flexible, hardy, strategy to deliver the returns. Why? Because specialisation is the mother of death!

When one day, the underlying fundamentals change as for sure they will, you will be gasping for a hardy, flexible strategy[a la Swiss Knife] and not a lean, mean, high specialised hack saw blade.The idea being,when one of the profiting markets turn mellow, we can “hope” that one of the earlier non profiting markets pick up and return some of the lost luster and sheen.

We often make decisions based on how much dollars is being returned. Thats a wrong move. You should be instead looking for a homogeniety in the results. A band of acceptability with none of them, sticking out.Probably some might give upside returns of 1.5-2σ away from the average. There is nothing to be really emphatic about. Your system will soon mean revert, and that part where you will try to lock in profits, will be handled by the money management part. For example strategies like Exposure Reduction, Staying Out of the Market, Equity Curve Trading etc etc.Because all we have in our hand is to reduce our drawdowns and manage our risks.

Till the next time,
Trade Well,

One Comment leave one →
  1. July 7, 2009 11:09 am

    Great site! Very professional looking. Thanks!

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 )

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

%d bloggers like this: