Kevin_in_GA
4,599 posts
msg #100733
- Ignore Kevin_in_GA |
5/14/2011 1:55:22 PM
Bill:
You're comparing apples to oranges here. The analysis I did was based on 2 years of VIX data, and not on any ETFs.
Did you actually try to buy and sell the Q's based on the VIX signal or on the Q bollinger bands? The former case is what I had tried to analyze, not the latter.
|
wkloss
231 posts
msg #100734
- Ignore wkloss
modified |
5/14/2011 2:59:57 PM
Kevin wrote:
"You're comparing apples to oranges here. The analysis I did was based on 2 years of VIX data, and not on any ETFs.
Did you actually try to buy and sell the Q's based on the VIX signal or on the Q bollinger bands? The former case is what I had tried to analyze, not the latter."
I was using Q bollinger bands for the Q test. I also tested 8-10 ETF's using their bollinger bands (not the Q's bollinger bands) @ (9,2.0) and the results for the past 2 years were amazing especially since I included some inverse ETF's. My surprise came when I tried to use the Q's beyond 2 years back.
I would have used the VIX signal to backtest trading the VIX but I didn't think that could be done on SF. Can it be and can I use the VIX signal to test ETF's ?
Thanks for helping me get on the right track. Your research is always interesting.
Bill
Update:
I did additional backtesting using SPY going back in 2 year increments. The results are:
BB (9,2.0)
PRICE ABOVE UPPER BB
5/15/2011 5/1/2007 4/30/2009 33% 67% 0.50:1 0.57 -20.27%
PRICE ABOVE UPPER BB
5/15/2011 4/29/2005 4/30/2007 31% 69% 0.44:1 0.63 -6.73%
PRICE ABOVE UPPER BB
5/15/2011 4/29/2003 4/29/2005 46% 54% 0.86:1 1.89 12.24%
My initial reaction was that something caused this system to work only from 2009 to 2011 but that doesn't make sense. You didn't test this on SPY but since the VIX is derived from SPY options, SPY seems like a good choice. My expectation before backtesting was that different volatility ranges would still work but gains would be smaller.
I suspect that the concept remains sound but that entry and exit rules should vary with the current range of volatility.
Your thoughts?
|
four
5,087 posts
msg #130043
- Ignore four |
7/25/2016 11:52:48 PM
pop
|
BarTune1
441 posts
msg #130102
- Ignore BarTune1 |
7/27/2016 10:19:13 PM
Still here and still relying heavily on the $VIX to time long/short entries ...
|
SAFeTRADE
644 posts
msg #130113
- Ignore SAFeTRADE |
7/28/2016 6:25:35 AM
My 2 cents on this very good filter. So simple.
|