Live Chat Email Us Call : 011-42563333

How to test a trading system.?

Mind Without Fear, asked a question on trading systems :
Quote
I am trying to answer a slightly different question and I seek your help in trying to figure out what could be the right approaches to address it.
You see I am using essentially a 2MA difference system. Back testing suggests it is profitable over the 6 month period.
I am wiling to try it out for myself at this point.And that brings me to the question. How can I set up a process to detect whether the 2MA system will still be relevant over the next year or two years?
In other words, If I assume that the system that worked for the last six months might degrade over the next year, how can I detect such a degradation as early as possible?
UnQuote

Before actually using a trading system, it should undergo a test. I created a video on this topic with a commentary, (the video cannot be seen since google rendered it in a small frame, but the commentary is easy to understand).

The main question was: how to detect a degradation in the system ? The way to do this is: determine the three or four important parameters for the system. Identify a cutoff point for these parameters. In actual trading, if the system begins to go below the cutoff points, you have to relook at the system and accept the fact that the environment may have changed. One way is to calculate the maximum loss incurred till date. In actual trading, if the loss exceeds twice the recorded maximum, then a relook is necessary.

Markets: What the world is saying

Citibank analysts in their European Strategy team say:

Markets in Twilight Zone, may have bottomed out.
Quote
2009 is shaping up to look more like a twilight zone. Earnings are falling faster than share prices, the market is re-rating, cyclicals are re-rating aggressively and earnings momentum strategies are struggling — all signs of twilight zones. Are we saying that the next bull market has started? No, but we are saying that markets have stabilised and are unlikely to fall beyond the March 2009 lows. Earnings declines have further to go from here. So with flattening prices and falling earnings, we think this is the start of a protracted twilight zone. It is sensible to gradually increase exposure to risk through the year. But, near term we would be less willing to chase the current risk rally.
UnQuote

But the other point of view is this:

Clusterstock says: The bear market has much more to go.

Quote

Now that stocks have rallied nearly 30% off their low, pundits agree: It’s a new bull market. So be very afraid.
Market punditry is a lagging indicator, not a leading one. Pundits are excellent at describing what has happened, not what is going to happen.
But doesn’t the 30% rally off the bottom obviously mean that the bears are fools, that it’s finally safe to get back in the water? No. It doesn’t obviously mean anything.

UnQuote

A chart of the previous three major bear markets is given in the site. These were: the 1929 bear market in the USA, the 1989 bear market in Japan and the ongoing 2000-todate bear market in the Nasdaq. The chart says that the markets remained below their bull market highs, even after 10 years. The price action was essentially sideways for most of this period.

My Notes: While we cannot say what will happen in he next 10 years, it is very unlikely that a new bull market will start suddenly.
Factors working against a new bull market starting Now:
The NSE website tells us that the trailing P/E for the Nifty on 29-Apr-2009 is 16.53. This is not a bargain level. With earnings likely to remian under perssure, any increase in share prices will push the PE up. That is high risk.
GDP gowth is expected to be between 4.5 and 6 percent this year 2009-2010. Not enough to meet the bare requirements of a growing population.

But the markets can go up anyway.
Traders must go with momentum, follow the trend. If it is up, we buy. But, we must understand that the rally (if it comes about now) will be a big bubble. So ? This is what I plan to do:
In my overall portfolio (this is about investing..), increase equity allocation only on sharp declines. Take profits on part of the portfolio if markets show a range expansion.
The risk in following such a strategy is to have lower gains if the current rally continues for ever, never giving a sharp correction. This is a low risk for me, since markets will always correct. The bigger risk is this: Sometimes markets continue to move up, without any pause, thus making many investors feel ‘left out’ in the rally. The investors then cannot bear the pain of being left out, and end up entering the market almost at the top. You need a calm mind to avoid falling in the ‘left out’ trap.

Leave a Reply

Your email address will not be published. Required fields are marked *