Rank: Advanced Member
Groups: Registered, Registered Users Joined: 4/16/2006(UTC) Posts: 39
|
It would be nice as you could test a system on a goup of stocks weekly or monthly data only. This by combining explorer functions and a systemtester in one function!
You might think at momentum system where you buy each first of the month the 10 stocks with the greatest average momentums in the previous 5 months. (use momentum filter in explorer and sort). Then use the explore to look each month if the stock is still in the top 10 (or any Top X nr. you fill in). If the stock is no longer in the top X, you sell that stock on the first day of the next month and replace that stock by the new coming stock in the top X. (systemtester function combined in explorer)
This system proves to be a very reliable winner technique at least this year but I would like to test this option rethrograde in time.
It is another way of testing trading systems.
|
|
|
|
Rank: Advanced Member
Groups: Registered, Registered Users, Subscribers Joined: 7/25/2005(UTC) Posts: 1,042
Was thanked: 57 time(s) in 54 post(s)
|
Anton
The April issue of MSTT has an excellent article by Harry Ward on creating an ROC ranking for mutual funds and holding the top five ranked funds (on monthly data). Funds entering the top five are bought and funds leaving the top five are sold. You can rank up to the 18 highest (or lowest) values by almost any criteria you might wish to use.
MetaStock does not make it easy to rank securities such that they can be back-tested, but by using my ranking tools with Trade Equity explorations or the Enhanced System Tester, it's not only possible, but Harry is already doing it.
Better still, you can use the EST or Trade Equity to create a portfolio equity curve as a by-product of your testing.
So in answer to your question, yes, there is another way of testing.
Roy
MetaStock Tips & Tools
|
|
|
|
Rank: Advanced Member
Groups: Registered, Registered Users Joined: 4/16/2006(UTC) Posts: 39
|
Thanks Roy,
I tested it on mutual fonds by hand in a free portfolio en quotes retrieving program (Stockbrowser). But the best profit it gave on single stocks and not on funds. Probably because you crop the fastest growing stocks (mid and small caps) so you can beat the index with more ease. (it beated the Dutch AEX/AMX index from 1-1-2005 till now each consequetive month with a about 100-200% and only 4-5 trades at the beginning of each month. This was without re-investing the gained profits and earnings). Would be nice to test it on the Morningstar Engine :D
Mo(C,20)+Ref(Mo(C,20),-20)+Ref(Mo(C,20),-40)+Ref(Mo(C,20),-60)+Ref(Mo(C,20),-80)-500 and a column filter for stocks with a momentum >100 for each period (or only the first 3-4 periods) and sorted upon this formula.
This formula does the same but catches off one day flies with a short but temporarily very high momentum period, from showing up too highly ranked:
(Sqrt(Mo(C,20)-102)+Sqrt(Ref(Mo(C,20),-20)-102)+Sqrt(Ref(Mo(C,20),-40)-102)+Sqrt(Ref(Mo(C,20),-60)-102)+Sqrt(Ref(Mo(C,20),-80)-102))*5
You need bullisch markets or you should change the momentum filter to <100 en look for the lowest numbers in bearish markets of course (for the visitors of this site). I am very curious how Harry Ward did his trick in Metastock. Do you have some more info for me on this subject (using the systemtester on monthly data and explorer results in one run?)
Anton
|
|
|
|
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.