Example: How to make an mechanical system.

Next step: determine risk and how much you are going to risk in each trade aka money management.

Before I begin, there are endless combinations, but for this example I'm going to use a straight forward approach:

On the daily graph you have an entry at $107.40 and a first initial stop at $112,19. This gives a risk amount of $4.79. The trade is stoped out at $100.71 due of the trailingstop. Profit on this trade is $6,69 or 1,39 times the amount you have risked on this trade. I'm not yet going in to how much capital you have or how much shares you must buy. That question will be answered later on, when we know more about our system.

On the Intraday graph you have an entry at $87,60 and a first initial stop at $87.30. This gives a risk amount of $0,30. The trade is still going at this moment, but the trailing stop is raised to $95,17. If we asume that we will not have a gap against our trade, the locked in profit is $7,87 or 26,23 times the amount you have risked on this trade.

Any questions so far?
 
The trade of the intraday chart has hit the stop of $95.17 and is closed out. This is of course all 'theoreticall' and does not account slipage and commision... But in real life trading you have to take this in account of course. Also notice that you have a new signal intraday later on in the chart with a new initial stop. At the end of the day the trend system is in a no-no state due that the 20 moving average is under the 50 moving average. When something of your system gets a 'negative' reading, just close out al trades. Do not 'hope' for it to change back, just follow the rules you have made for your trend system...

Tomorrow I will go further with money management...
 

Attachments

Before I go on money management, I will show how important it is to use a market filter for any system.

The money management calculation I have used in this example is to risk 1% of the equity($25000) in every new trade. Pretty simple...

I have programmed the system for daily bars in Mechanica to let you see how to implement the filter and why. This could be an eye opener for some people that are trying to design there own system:

First, you see an equity with all the years with no marketfilter. The marketfilter we are going to use is a 9 period moving average on a monthly chart. When the 9 period is up from the prior period we implement the system, when it is down, we do not implement the long trend following system. No shorts for now...

Equity char all years without marketfilter:
 

Attachments

You can see that the equity chart was in a drawdown from '88 untill '98. This is because AAPL, our test symbol, was not realy in a 'bull' mode in that period. You maybe will not notice because of the logaritme(or lack of it)scale of the equity chart, but also in that drawdown period there where some periods that the equity went up... Now for keeping it simple let's go to a point where there was a clear uptrend of the stock according our filter. The beginning of '98.

The next equity chart shows the equity from '98 untill now. Including the bear of 2000.

Notice the long 'flat' period of the equity chart.
 

Attachments

You see that the long flat period is exaclty the same time where there was a 'bear' market. Now in our case it did not realy hurt our equity, but that is pure 'luck'. It could easy been a drawdown on your equity in that period.

Now, lets use our marketfilter. The first go period is from 98 untill oktober 2000. The second go period is from may 2003 untill july 2008.

Equity chart of the first go period:
 

Attachments

Quote from danielc1:

Now, lets use our marketfilter. The first go period is from 98 untill oktober 2000. The second go period is from may 2003 untill july 2008.
More...

Marketfilters ocan only do you any good on hindsight.
 
Equity chart of our second go period. I have put in the end capital of the first period to have a compound result in total dollars...
 

Attachments

You must log in or register to reply here.
Back
Top