Kevin_in_GA
4,599 posts
msg #113177
- Ignore Kevin_in_GA |
5/7/2013 11:19:12 AM
Stratasearch does this calculation without issue, unlike SF. I think this is a decent work-around, but one could also reduce the number of divergence filters from 10 to 5 (that all work as they should using SF code) and you'd be OK as well.
|
jimmyjazz
102 posts
msg #113185
- Ignore jimmyjazz |
5/7/2013 12:19:45 PM
Thanks. Is your 10-filter C2 system a composite version; i.e., you issue a single buy or sell for the index, or do you trigger partial trades based on the number of "agreeable" indicators?
|
Kevin_in_GA
4,599 posts
msg #113188
- Ignore Kevin_in_GA |
5/7/2013 1:56:31 PM
Each filter element controls 1/10th of your trading positions in SPY. Exits are when the exit signal is triggered without a corresponding divergence signal still in play. Sounds a little confusing, but it is pretty simple to do.
|
jimmyjazz
102 posts
msg #113189
- Ignore jimmyjazz |
5/7/2013 2:06:54 PM
Sounds like my accountant would have a stroke!
|
jackmack
334 posts
msg #113190
- Ignore jackmack |
5/7/2013 2:16:22 PM
Kevin
Not sure if you saw this but thought if anyone could do something with it - it would be you
http://www.etfscreen.com/blog/2013/02/26/monthly-seasonality-and-etfs/#more-1048
I know the divergence filter is killer but just thought maybe you could work on something like this to add to quiver of amazing filters.
Just a thought
Cheers
|
cnncurt
2 posts
msg #113556
- Ignore cnncurt |
5/20/2013 9:48:20 AM
Kevin -
With Pangolin D, are you using the composite filter (p. 22?), and if so what is your criteria for issuing buy and sell signals?
|
Kevin_in_GA
4,599 posts
msg #113558
- Ignore Kevin_in_GA |
5/20/2013 10:01:51 AM
With Pangolin D, my criterion for issuing BUY and SELL signals is that you are a subscriber.
|
miko
68 posts
msg #114388
- Ignore miko |
7/9/2013 10:28:26 PM
Was today an exit signal from all open positions?
|
Kevin_in_GA
4,599 posts
msg #114391
- Ignore Kevin_in_GA |
7/9/2013 10:49:17 PM
Yes. No current divergences are in play and the exit signal triggered. Note that Friday was also an exit day.
Just to let everyone know, I have re-optimized the divergence filters. They have been consistent winners (and will likely continue to be) but they were developed 10 months ago using a look-back period of about 5 years. I felt it was time to update them with the recent data. The new filters have higher win percentages and better Sharpe ratios, but are likely to trigger a bit less frequently than this set.
They are now the revised Pangolin D system on Collective2.
|
miko
68 posts
msg #114411
- Ignore miko |
7/10/2013 7:15:05 PM
Thx, somehow I missed the Jul 5 exit signal.
|