FTSE System, can someone help me test it please?

[Davidee]

I just did an SQL query to test around 10 years of OHLC FTSE Data from Yahoo in an Access database. I wanted to know what would have happened if a person went short when the FTSE 100 fell below the previous days low and went long each time the indice went above the previous day's high; exiting all trades when the market closed.

This produced a system that was 'right' about 63% of the time (counting all days where both the previous days highs and lows were broken as lossing days).

Can anyone help me test this 'properly' please?

I don't imagine this will work on the Dow, probably because of the way it is calculated. But it *might* work on the FTSE.


Thanks

 

[donaldduke]

How were you testing gap openings, where the ftse opens higher or lower than either of yesterdays extremes?

 

[Hoggums]

I just did an SQL query to test around 10 years of OHLC FTSE Data from Yahoo in an Access database. I wanted to know what would have happened if a person went short when the FTSE 100 fell below the previous days low and went long each time the indice went above the previous day's high; exiting all trades when the market closed.

This produced a system that was 'right' about 63% of the time (counting all days where both the previous days highs and lows were broken as lossing days).

Can anyone help me test this 'properly' please?

I don't imagine this will work on the Dow, probably because of the way it is calculated. But it *might* work on the FTSE.


Thanks

I tested this very system probably about 2 years ago - it doesn't work. The problem is the close for the day is often in between yesterdays high/low - so even if your trade opens it will probably end up negative by the end of the day.

 

[Davidee]

Thanks

Best that I'm told now that my idea sucks and learn the leason now, rather than learn it the hard way.

 

[Ursssk]

Don't lose faith dude. I work with a system that trades breakouts above previous price extremes and it makes me money.
I wonder what would happen if you varied the period of time from which the high/low is measured - ie what if you traded a breakout of a 3 day high/low? In terms of testing I can't help I'm afraid, but hopefully someone could shed a little light?

 

[mr_cassandra]

one thing I've learned over the years is try and be able to run your own analysis. Whether its in access, excel, wealth-lab, amibroker, whatever. You can not trust all the input you will receive on internet boards as actually having done the homework. Some of them do, but some are merely 'armchair system critics', never having set foot on the field.

I was told with great assurance years ago that if vix was any good as an indicator, everyone would start using it and it would become self-negating.

7 years later, the pontificating guru is gone and vix is as useful today in my program as it was many years ago.

Listen to everything presented but be able to check on your own. Machines don't have emotions or egos to protect and if the program says your idea loses money, so be it. If you had your ftse data in excel, you could copy parts of my sample program and use it to make buy-sell-loss-gain logic in your idea, then seel the real results, all without conjecture or assumption. By so doing you can assess the opinions you get.

my yahoo group is listed on my profile link, there is a free program download in the files section