logo
Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Options
Go to last post Go to first unread
Liadan  
#1 Posted : Monday, September 8, 2008 9:16:08 AM(UTC)
Liadan

Rank: Advanced Member

Groups: Registered, Registered Users, Subscribers
Joined: 6/12/2007(UTC)
Posts: 459

Use a Points Only Test When Running a System Test on a Commodity.

The gain/loss amounts of a test are based upon the equity invested and the price change of the security. The true gain/loss for a commodity trade is based on the dollar value you earn for each point you gain or lose in the commodity and the dollar value each point change is worth and the number or contracts you purchase. A system test does not know the dollar value of each point or how many contacts you purchased so the gain/loss amounts will not be the true dollars gained or lost for the commodity trade.

So when you run a system test on a commodity, you will probably want to run a Points Only test. A points only test will calculate the number of points you gained or lost when you bought and sold, using your trading rules. Once you know the number of points you gained or lost you multiply the value that you know each point move is worth, by the number of points gained or lost to determine for yourself the amount of money the test would have made or lost for you.

AliceOB  
#2 Posted : Thursday, January 20, 2011 10:49:52 PM(UTC)
AliceOB

Rank: Newbie

Groups: Registered, Registered Users
Joined: 1/21/2011(UTC)
Posts: 2

I am trying to do such a back test. I run it as a points test and I get results showing the greatest to least points over the time period specified. Can this be reported as % gain? e.g. gold could move from 1100 to 1400, eg a 300 point gain or 27%; while the Hang Seng could move from 22700 to 24700, eg a 2,000 point move but it is actually an 8% move - so gold would be more profitable. Say I have $100,000 to invest, I split this 10 ways of $10,000 each. so for gold I would place $9 per point (10,000/1,100) and for the Hang Seng it would be $0.4 per point (10,000/22,700), so the most profitable is the one that moves the greater %, not the number of points - and this is what I would like to back test. Any suggestions on how I can do this? I have tried exporting the results to excel to include the commodity value and calculate the % gain from there, but you would have to lookup the commodity value for each buy signal to get an accurate report.
seppia  
#3 Posted : Sunday, February 27, 2011 11:14:54 AM(UTC)
seppia

Rank: Newbie

Groups: Registered, Registered Users, Subscribers
Joined: 2/15/2011(UTC)
Posts: 1

The Points Only test present some problems. For instance: if I test my trading rule on different kinds of Commodities I cannot use the Optimizer, because the overall results are in points and not in dollars, so the optimizer simply sums points with really different values (thinks also to Futures in other currencies). The worst of it all is that you cannot set the number of contracts: in the real life this is very important due to: broker's conditions (especially for the margins requested), and risk management. Lastly some very useful third party add-on like JBL doesn't work.
When MetaStock adds a very simple feature like a conversion table for the System Tester (not on the original data as now) ? Something like: currency, value, # contracts.
Users browsing this topic
Guest (Hidden)
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.