duke56468
683 posts
msg #111460
- Ignore duke56468 |
2/17/2013 6:26:33 PM
Mike... Kevin can much better tell you how it is supposed to be traded, but what I tested was to enter SSO the first time a minimum of 2 divergence filters triggered on the same day, then exit when it said exit (I used SSO in the filter instead of SPY). Sometimes I had to enter the next day after I had exited. If I were in cash and had a simultaneous entry signal and exit signal I stayed in cash.
|
mahkoh
1,065 posts
msg #111465
- Ignore mahkoh
modified |
2/18/2013 6:06:50 AM
What you could try is to allocate 10% of the amount you wish to put at risk to every filter.
I have come to like the following indicator
Be cautious when it looks to be topping.
The filter was optimized for SPY, obviously SSO is connected but you are dealing with decay issues if you are forced to hold for an extended period. You could choose to short SDS instead of buying SSO.
Kevin already mentioned that the bulk of profitable trades are closed out within 6 -10 days, you could incorporate a max holding period.
|
sohailmithani
192 posts
msg #111478
- Ignore sohailmithani |
2/18/2013 10:25:18 PM
I did mention this issue with implementing this strategy regarding these huge losses in few of the trades. I also tried 10 days as maximum hold days and backtested that straegy as well. The results are getting worse as some of the good trades are hitting stop losses criteria too. I will keep trying other measures to mitigate this risk. I will share with the team if I will find anything useful.
|
duke56468
683 posts
msg #111514
- Ignore duke56468 |
2/19/2013 12:15:21 PM
Kevin.. I know this is an EOD filter, but I was wondering if it is prudent in your experience to exit intraday if the filter exit triggers (like today). I realize it is personal judgment, just wondering if you had an opinion about that. We will see how it works out.
|
Kevin_in_GA
4,599 posts
msg #111522
- Ignore Kevin_in_GA |
2/19/2013 4:12:53 PM
I have not looked at intraday entries or exits. No good way to backtest them.
Note that with today's exit trigger the 9 filters currently in trades will once again all close profitably, making the results since December 2012 103 winning trades, 0 losing trades. Not bad for a purely mechanical system. People can tweak this all they want, but I doubt that they will meaningfully improve the results.
I might also suggest that people stick with SPY instead of simply using SSO to get some leverage. They are related but different, and give different results. So when I see people saying "I traded SSO and it does not act the same" I think "and why did you do that? That's deviating from the system, and you knew that when you placed the trade."
|
wkloss
231 posts
msg #111524
- Ignore wkloss |
2/19/2013 4:44:32 PM
I have been trying to follow the posts on this system and I am getting confused.
How could you have 100+ trades since December 2012. There haven't been that many days. Are you saying 100+ signals and if you took all of them, the result would have been 100+ winning trades?
Bill
|
novice07054
26 posts
msg #111527
- Ignore novice07054 |
2/19/2013 5:57:04 PM
Please paste the correct filter as there have been modifications.
|
Kevin_in_GA
4,599 posts
msg #111529
- Ignore Kevin_in_GA |
2/19/2013 7:22:33 PM
There are 10 separate filters here, each one independently generating BUY signals but all using a common exit signal. This was all clearly described in the first post.
From these 10 filters 103 separate BUY signals have been generated, all of which resulted in winning trades.
Hopefully this clarifies things.
As to modifications to the filter set ... There have been none from me. Use either the separate filters I posted on page 1 or the composite filter on page 2.
|
novice07054
26 posts
msg #111530
- Ignore novice07054 |
2/19/2013 7:39:43 PM
As always thanks and fabulous work.
|
sohailmithani
192 posts
msg #111531
- Ignore sohailmithani |
2/19/2013 8:12:41 PM
Hi
When I ran the filter as posted in one of your email threads I get the following results;
Thursday February 14, 2013
Symbol Last Chg(%) total buy sell
SPY - SPDR S&P 500 ETF 152.29 0.09 7 1
February 15. 2013
[download csv]
Symbol Last Chg(%) total buy sell
SPY - SPDR S&P 500 ETF 152.11 -0.12 6 0
Tuesday February 19, 2013
[download csv]
Symbol Last Chg(%) total buy sell
SPY - SPDR S&P 500 ETF 153.25 0.75 2 1
As per Kevin's ;ast email which states that there were 9 open trades, can somebody please clarify to me as to which filter is being used here? If I am reading the data correctly I should have had 6 open trades that showed up on Feb 15th since we got the sell signal on Feb 14th. Am I using the right filter? or missing something here? Since I am new to this discussion and trying to follow up on this strategy, any help in this regard would be appreciated.
|