jackmack
334 posts
msg #112732
- Ignore jackmack |
4/12/2013 12:55:18 PM
Kevin
Just curious
Your post #112595 from 4/2/13
You stated "Having backtested this new approach, it works quite well"
Which version were you referring to - the new combined filter from Cheese or the spreadsheet version and the combined entry criterion from frsrblch.
Just curious
Thanks
Cheers
|
Kevin_in_GA
4,599 posts
msg #112733
- Ignore Kevin_in_GA |
4/12/2013 2:16:39 PM
I was referring to the approach where if you recieve both a BUY and SELL you stay in the trade. The original stats I posted for each filter had you exit the next day and then immediately rebuy - at what should be the same price, but you have to add in commissions and possible slippage.
|
jackmack
334 posts
msg #112734
- Ignore jackmack |
4/12/2013 2:28:23 PM
Ah - got it - I misunderstood the statement.
Thank you for the clarification.
Cheers
|
frsrblch
35 posts
msg #112736
- Ignore frsrblch
modified |
4/12/2013 3:54:34 PM
You can fiddle with the entry and exit criteria in my spreadsheet (the column with blanks and ones immediately to the left of the entry and exit price columns) to test different approaches, but my testing seemed to show that holding when you get both a buy and a sell signal on the same day appears to be the best approach.
|
mahkoh
1,065 posts
msg #112737
- Ignore mahkoh
modified |
4/12/2013 5:00:31 PM
Taking a closer look at the exit brings up a question: In your "Exit" column in the Exit tab the statement is =IF(AND(F2=1,F1<>1),1,"") which means that if there has been an exit yesterday there can not be an exit signal today. Why is that?
I can't find any reference to that cell anywhere.
|
frsrblch
35 posts
msg #112738
- Ignore frsrblch
modified |
4/12/2013 8:22:00 PM
The G column in the Exit sheet isn't actually used. Originally, I had built the sheet around buying or selling when the signal goes from 0 to 1, regardless of what the other signal is doing, instead of the (entry=1,exit<>1) and (exit=1,entry<>1) it uses now. I changed how the exit was calculated at some point, and forgot to delete the unnecessary column.
|
jackmack
334 posts
msg #112742
- Ignore jackmack |
4/13/2013 5:31:44 PM
Kevin
I was just thinking about your TAA system and how you move into differing funds according to set parameters end of month.
Then I began to wonder if it would be possible to somehow work the that methodology into this filter set and "maybe" yield a potentially better return over the straight SPY set?
Meaning if say IWM were pulling ahead using the TAA system you moved out of SPY into IWM would it be possible that there are divergences in IWM that would allow better returns for said month vs. SPY throughout?
Kind of a TAA DIVERGENCE filter?
Or are divergences not present in the EFA - IWM - MDY as they are in SPY.
Just wondering?
Thanks
|
jackmack
334 posts
msg #112765
- Ignore jackmack |
4/15/2013 10:25:34 AM
On second thought - no way that would really work
Oh well - no harm no foul for thinking out loud
Thanks anyway
Cheers
|
Kevin_in_GA
4,599 posts
msg #112766
- Ignore Kevin_in_GA |
4/15/2013 11:16:25 AM
These divergences are unique for ^SPX.
|
mahkoh
1,065 posts
msg #112768
- Ignore mahkoh
modified |
4/15/2013 11:32:12 AM
You could translate it to ^RUT this way
But as Kevin stated you'll see that this does not act the same on entries and exits.
You would need to re-optimize the whole filter. I actually have been thinking about that as this means you can go short TZA on a signal. TZA's decay could boost this filter's win rate to over 90%.
Note that the CMF filter will also not trigger here.
|