Kevin_in_GA
4,599 posts
msg #92869
- Ignore Kevin_in_GA
modified |
5/18/2010 8:43:02 AM
Is there a way for SF to backtest this?
VXX has only been in existence about a year, and the total number of trades based on crossing the MA(13) is less than 20. Should be easy to do manually using the filter I posted earlier in this thread.
|
mesayah
658 posts
msg #92985
- Ignore mesayah
modified |
5/21/2010 11:58:05 AM
Kevin. I really didn't care for your work but from the way this looks this may have propelled
you to the top of the class. It would be interesting if you could figure out and post how BGU-BGZ would have performed over a time period based on %.? It seems VXX performed great.
|
mesayah
658 posts
msg #92992
- Ignore mesayah |
5/21/2010 12:38:41 PM
So would you consider trading this exclusively and dump all other investments and why would you short VXZ over VXX?
Do you think it's difficult to short them?
|
Kevin_in_GA
4,599 posts
msg #92995
- Ignore Kevin_in_GA |
5/21/2010 12:55:23 PM
So would you consider trading this exclusively and dump all other investments and why would you short VXZ over VXX?
Do you think it's difficult to short them?
+++++++++++++
No. I would add this into whatever trading strategy one is currently using - this is an opportunistic play. It is hard to short VXX since everyone is trying to do it over the past year.
The suggestion to short VXZ and go long VXX is a classic pair trade hedge strategy - it will reduce your overall return but will also more dramatically reduce your volatility and risk of large drawdown.
I look at this as both an indicator and as a trading play - one could simply go long on the SPY using UPRO until the VXX:SPY ratio crosses the moving average, then flip over to VXX. This has worked quite well over the past year but that is mostly due to relentless QE that kept propping up the markets at the expanse of volatility.
|
duke56468
683 posts
msg #93003
- Ignore duke56468 |
5/21/2010 2:15:52 PM
The chart for the vxx:spy ratio looks nearly identical to the chart for vxx except for the numbers in the y axis. What is the thinking behind using the ratio 13MA and not just using the vxx 13MA? Is it to facilitate creating the filter?
The vxx 26MA 52MA cross which just occurred 5/18 did not produce any whipsaws in the last year.
|
Kevin_in_GA
4,599 posts
msg #93006
- Ignore Kevin_in_GA
modified |
5/21/2010 2:28:30 PM
My thinking is that the ratio will be more sensitive to changes, because as the SPY goes up, the VXX goes down and the ratio is affected in both numerator and denominator.
No reason not to look solely at VXX. I just like to have both sides of the trade represented. Normally the two elements in the ratio are positively correlated (example, OIL and XLE), but they don't have to be for this to tell you when it might be a good time to go long volatility versus the overall market.
|
mesayah
658 posts
msg #93014
- Ignore mesayah |
5/21/2010 4:27:14 PM
VXX will go to 63
|
stoxrox
10 posts
msg #93062
- Ignore stoxrox
modified |
5/23/2010 11:45:03 AM
Hi Kevin,
I noticed the $vix:spy is beginning to show bullish divergence on the CCI(14) and also on the MACD histogram. Maybe we are in for a continuation of Friday's pop. What are your thoughts?
|
mesayah
658 posts
msg #93065
- Ignore mesayah |
5/23/2010 2:12:31 PM
Kevin you have the Big Bonanza winner Filter here in trending environments.
Can you extend the courtesy as you reamed me for and tell us how it's done
% wise since inception of VXX or thru many market environments? I want to
trade your indicator exclusively. Can you help??
Oh and can you refine it anymore for choppy markets??
|
Kevin_in_GA
4,599 posts
msg #93067
- Ignore Kevin_in_GA |
5/23/2010 2:57:09 PM
Kevin you have the Big Bonanza winner Filter here in trending environments.
Can you extend the courtesy as you reamed me for and tell us how it's done
% wise since inception of VXX or thru many market environments? I want to
trade your indicator exclusively. Can you help??
Oh and can you refine it anymore for choppy markets??
+++++++++++++
VXX has only been in existence since Jan 2009. Can't backtest any farther than that. I provided a simple filter that can be manually backtested in the forth post on this thread. I would use this filter in the following manner:
1. Always long the SPX when this ratio is below the MA(13).
2. Buy VXX when the ratio is above the MA(13) - if you look at VXX perfromance during these times, it easily beats even the 3x Bear ETFs.
I will manually backtest this (and encourage others to do the same with other stocks such as TNA, UPRO, etc) but am travelling on business for the next few days.
|