Kevin_in_GA
4,599 posts
msg #93540
- Ignore Kevin_in_GA |
6/5/2010 9:07:45 AM
Yes thanx Kevin. Down 45% since the tout 2 weeks ago...
+++++++++++++++++
The VXX high over the past 3 months occurred on 5/21 at 36.22, and the highest close occurred on 5/20 at 34.07. The VXX closed on Friday at 31.29. By my calculation that only puts it down 13.6% from its high (which you never would have been able to time for selling), and only down 8.16% from its highest recent close.
Thank the good lord i did not buy into this. Is this what you meant by being responsible?
+++++++++++++++++
Had you bought in at the signal crossover on 4/27, when VXX was at 20.50, and sold on the return crossover on 5/27 when it closed at 28.33, you would have made a tidy gain of 38% in a month.
If you traded on the signals, the recent performance has been better than most people would be able to claim for any other system they used over the same timeframe.
This week we saw a whipsaw - in at 30.19 on Jun 1st, and then out at 28.87 on Jun 2nd for a loss of 4.3%. No system is immune to these, especially given the recent market action. That means that if you had bothered to invest $10,000 at the start, then re-invested the entire gains ($13,800) into the whipsaw, you are now at $13,206, still up 32% from five weeks ago.
This is why i don't like to give all my trade secrets away. The system seems to stop working,
+++++++++++++++++
Again, you are being a little quick to judge - the VXX:SPY ratio whipsawed over the past week, but is now back above its MA(13). Let's see where the next week or two goes before eulogizing this filter.
That being said, a backtest of this ratio leads to more losing trades than winning ones, simply because of the relentless downward slope of VXX over its short existence. The real winners have been to go long on the SPX when the ratio is below the trigger line (the OTHER half of this approach, which no one has mentioned yet).
I believe this indicator can be helpful, but with such a runup that occurred in such a brief period of time in both SPY-VXX
that whipsaws were inevitable.
++++++++++++++
Agreed. The markets are highly volatile, and no single system will always work. The real test of any system is returns over time - since VXX has only been around for a little over 16 months, there is not a large amount of data for backtesting.
|
mesayah
658 posts
msg #93543
- Ignore mesayah
modified |
6/5/2010 10:27:52 AM
What's at question is if a system stops producing at some
point either on it's own accord or from too many using it?. And how
would you know to scrap a system, especially an indicator unless you
have taken several big hits. And why should you share if it reduces the
value of the system. You're basically shooting yourself in the foot.
Catch 22, for you will never know when to bail.
|
Kevin_in_GA
4,599 posts
msg #93546
- Ignore Kevin_in_GA |
6/5/2010 12:11:56 PM
What's at question is if a system stops producing at some
point either on it's own accord or from too many using it?. And how
would you know to scrap a system, especially an indicator unless you
have taken several big hits. And why should you share if it reduces the
value of the system. You're basically shooting yourself in the foot.
Catch 22, for you will never know when to bail.
+++++
Too many people? Using this filter? Not really a concern. Nothing the combined might of every SF member, all trading the same way at the same time, could do to change anything on the VIX or major markets.
In fact I believe there is a commonly held misconception about the power of retail investors to influence the markets. We do not have much of a say in how stocks or markets trade - the big institutional investors do.
My philosophy is that sharing trading ideas leads to better outcomes, not worse. I learn as I go, and am always willing to share my observations.
|
mesayah
658 posts
msg #93907
- Ignore mesayah |
6/16/2010 11:48:45 AM
Kevin... I think you should just go ahead and
manually set the change to VXX. It's time
to go short...
|
Kevin_in_GA
4,599 posts
msg #93911
- Ignore Kevin_in_GA |
6/16/2010 12:21:18 PM
The filter is what it is - it only tells you one perspective. Right now the ratio has dropped another 1.42% as the SPX holds at 1115. So far it says to stay long. Your call as to whether or not you agree.
I actually noticed that using the ROC(12) above 0 on this VXX:SPY chart works fairly well also - actually avoided some whipsaw action recently. I did not really look closely at it when I put this chart together, but it may have some better predictive value than the MA cross.
|
richo
74 posts
msg #93945
- Ignore richo |
6/17/2010 6:34:10 AM
Kevin,
On the ROC(12) are you saying above 0 go short and below 0 go long? Thanx for your contributions
Rich
|
Kevin_in_GA
4,599 posts
msg #93951
- Ignore Kevin_in_GA |
6/17/2010 9:57:22 AM
richo:
Basically yes. Look at the link to the Stockcharts.com chart of this ratio in the first post. You can see that the ROC generates a more stable signal than the MA crossover - but I have not backtested this to see if it actually performs better.
I am trying to figure out how to get SF to recognize the ROC function on a ratio. I may have to code it myself just to see.
|
student
14 posts
msg #93985
- Ignore student |
6/17/2010 6:06:42 PM
Kevin: How did you get VXX:SPY and $SPX on the same chart? Is paid membership needed at stockcharts.com to do this? Thanks.
|
richo
74 posts
msg #93999
- Ignore richo |
6/18/2010 6:38:05 AM
Thanks for your response Kevin it is much appreciated
Rich
|
Kevin_in_GA
4,599 posts
msg #94255
- Ignore Kevin_in_GA |
6/24/2010 7:08:25 PM
Well, looking at the chart, the VXX:SPY ratio has once again crossed above its MA(13).
This does not bode well. If tomorrow does not have a reflex "bounce" after 5 down sessions in a row, we are heading for a real rough patch.
Look for a gap up if you want to hedge your portfolio with VXX, TZA, or SPXU.
|