Kevin_in_GA
4,599 posts
msg #111323
- Ignore Kevin_in_GA |
2/12/2013 1:42:41 PM
@sohail: Please look at the performance metrics in the first post. This system was not tested against 2002 data, but in the period from 1/2/2007 until recently it crushed the market. It would be hard to say that this did not perform well in 2008 when the results are typically 85-90% wins.
@Miketranz: SPY is at 152 and change.
|
radiobb2
10 posts
msg #111330
- Ignore radiobb2 |
2/12/2013 5:13:23 PM
Well, I got in today, but now it's saying get out. Seems like it trades too much in the past week.
|
Kevin_in_GA
4,599 posts
msg #111331
- Ignore Kevin_in_GA |
2/12/2013 5:45:36 PM
If every one of the last 60 trades is a winner, how much is too much?
So you got in today at 151.78 and technically you don't exit until tomorrow at the open. But 9 of the 10 divergence filters are still signalling a BUY. Your call on what to do ... the market might react negatively to tonight's State of the Union address and the system will record it's first set of lossess since December. Or you can hold pat and use the fact that the signals are still in BUY mode and see how the rest of the week goes.
|
radiobb2
10 posts
msg #111332
- Ignore radiobb2 |
2/12/2013 6:01:35 PM
Overall, it seems to work well. The only issue I have is avoiding trade settlement rule issues. In and out twice in a few days is a free ride. I'll plan to hold tonight and see what happens.
|
sohailmithani
192 posts
msg #111335
- Ignore sohailmithani |
2/12/2013 7:17:43 PM
Kevin, no doubt the success factor of 80% to 90% is a great job done. Thanks for your valuable contribution. I just wanted to know your thoughts regarding the bear market returns. When I ran the back test for the years 2002 and 2008 and analysed the data, I found that the success rate were still great. The issue was that the full year still resulted in negative returns only due to few trades that produced higher -ve returns in excess of -20%. Any thoughts ragarding the most effective stop losses that you would recommend to maximise our returns?
|
Kevin_in_GA
4,599 posts
msg #111339
- Ignore Kevin_in_GA |
2/12/2013 9:51:03 PM
The successful trades are almost always completed within a week or so. Consider a time-based exit of 5-6 trading days.
|
voidcomp
23 posts
msg #111347
- Ignore voidcomp |
2/13/2013 9:14:53 AM
Amazing market over the past few months. Up days to down days on SPY are nearly 2 to 1 and VXX continues to make new lows.
|
sohailmithani
192 posts
msg #111449
- Ignore sohailmithani |
2/17/2013 2:20:14 AM
Hi Kevin
After reading all 14 pages I feel confident and am planning to start trading on this strategy starting Tuesday. Please let me know if I want to follow the strategy I should buy 6 ubits of SSO / SPY on Tuesday? I am still a little bit confused on the sell signals that are coming up with the buy signals but will try to follow the posts going forward to have a hang of it. Your help would be appreciated.
|
Nurdish
2 posts
msg #111450
- Ignore Nurdish |
2/17/2013 8:40:50 AM
SF need to introduce a contribution recognition system (e.g like Reddit GOLD etc). The info provided by Kevin and many other posters is like $$$ in the bank.
|
Kevin_in_GA
4,599 posts
msg #111451
- Ignore Kevin_in_GA |
2/17/2013 10:13:25 AM
Statistics update - this last week has had a bunch of BUY and SELL signals on top of each other. Every trade closed positively but at very small gains (between 0.05% and 0.1%). In real trading these would probably end up as losses due to commissions and bid/ask spreads. That is why holding when a simultaneous BUY and SELL signal is probably a good idea.
Problem is I can't really code for that since it is a discretionary call. However, the system is 94 for 94 on recent trades using the backtest and strict adherences to the rules.
|