StockFetcher Forums · Filter Exchange · HELP FILTER 97% Success<< 1 2 3 4 5 ... 8 >>Post Follow-up
novacane32000
273 posts
msg #107327
Ignore novacane32000
modified
8/1/2012 8:34:39 PM

I manually backtested this on SF going back to 1/2011 and the below is what I found.

I went long XIV when RSI(2) on the VXX went above 90 and scaled in 1-2-3 units as mentioned.
I had 19 W and 2 L. Using $1000-$2000 and $3000 the 19 winners = $2001 and the 2 losers = -$141 . Add a digit for WOW ! effect

I also went long VXX when RSI(2) on XIV went above 90 but the results were not as good.
17 W and 9 L but for a net loss of $132 (using 1000-2000-3000). One huge loss of 17% was in March 2012 during an unmistakeable uptrend and another 11% loss during the 2011 xmas rally. Take out those 2 and you did okay. I also tried entering only after RSI(2) crossed 95 and that did improve results , mainly by reducing the 2 big losers.

Lesson learned-do not go long VXX in an up trending market or during seasonally strong Nov Dec.
If you do, then use an RSI(2) 95 cross instead of 90

I believe the XIV trade worked out so well because the market (SPY) gained 9.4% during the test period and I would say that the VXX trade may also work out much better in a down trending market which did not occur in this backtest period.

Thanks to BarTune for providing the basic filter and Jnafach

BarTune1
403 posts
msg #107368
Ignore BarTune1
8/5/2012 10:19:54 PM

I have been on holidays and still am travelling .... I have only played this system twice profitably since I learned of it .... I see that the VXX is signalling a buy now .... not sure I am going to be around a computer to play it - but if it drops again I may put in a limit order ... you are correct that the short version seems much more profitable than the long version ....

BarTune1
403 posts
msg #107633
Ignore BarTune1
8/15/2012 1:25:12 AM

The long VXX trade triggered an exit yesterday with the VXX around 11.75 ... if you had averaged in (1-2-3) your cost was around 11.83 for a loss of less than 1%. Of course, the long side of this trade has not shown near as good results as the short side which will hopefully be the next set up.

My limit orders did not fill while on vacation but i was able to pick up some VXX at 11.45 and 11.17 in the past few days so I ended up with a small profit.

drew9
171 posts
msg #107645
Ignore drew9
modified
8/15/2012 2:16:50 PM

Sorry I did not go back and review this thread. Here is what I use to pinpoint tops and bottoms of the VIX.

set{vix10dayma, cma(ind(^VIX,close),10)}
set{5pct,vix10dayma * .05}
set{plus5%,vix10dayma plus 5pct}
set{minus5%,vix10dayma minus 5pct}

draw ind(^vix,close) on plot vix10dayma
draw plus5% on plot vix10dayma
draw minus5% on plot vix10dayma
draw vix10dayma

This was devised by Kevin so I will give him major kudos for it. Just add the lines to your filter. The best trades are typically when the VIX is above its MA by >5%. Is it possible to revise the results to enter trades on the first day when the VIX in the above scenario turns down? I would love to see the results for XIV as it is easy to enter with huge volume and no issues trying to short it. It also decays in your favor (about 10% per month) vs VXX against you. Either way the results should be similar whether you choose VXX or XIV. The above is just for a visual representation but however you have to code it for the VIX turning down will still provide the needed result.

So for XIV we would want RSI(2) below 10 but do not enter the trade until the CLOSE of the day the VIX actually declines. For the exit, please try three scenarios: 1) RSI(2) above 50 with the exit at the next days open, and 2) Same but with RSI(2) above 70 and 3) Exit 3 days after the entry at the open.

Also I if you could run the entry for RSI(2) below 20, I believe the the results will be slightly improved due to possibly a few more entries.

SF only allows backtesting at the next open but it appears others in this thread have other options. The key here is entry at the close of the first day it turns down. I doubt the pyramiding will come into play here but this would allow for larger positions. Thanks and I look forward to the results!

novacane32000
273 posts
msg #107652
Ignore novacane32000
modified
8/15/2012 8:12:03 PM

Drew ,want to make sure I follow you here.
You want to short the VXX when it is above the plus5% line but only on the first day the vix10dayma line moves down.
The 3 month chart shows 4 x when the VXX was > plus5%.
3 of the 4 were clearly profitable, the one in May was profitable holding 5 days.

Fetcher[

symlist(VXX,XIV)
add column RSI(2)
and rsi(2) is above 0

set{vix10dayma, cma(ind(^VIX,close),10)}
set{5pct,vix10dayma * .05}
set{plus5%,vix10dayma plus 5pct}
set{minus5%,vix10dayma minus 5pct}

draw ind(^vix,close) on plot vix10dayma
draw plus5% on plot vix10dayma
draw minus5% on plot vix10dayma
draw vix10dayma
add column vix10dayma
add column plus5%
add column minus5%
add column ind(^vix,close) on plot vix10dayma}
]




drew9
171 posts
msg #107655
Ignore drew9
8/15/2012 10:28:14 PM

No was trying to follow the filter with RSI(2) below 10 or for another example 20. Would use the confirmation of the VIX turning down before the entry. So we would need both RSI(2) one day prior below 20 and VIX on the same day turning down. Entry at the close. This avoids catching a falling knife and awaits the confirmation. The next step then will be to find the best ETF to use but have the VIX confirm the entry. I have done preliminary work on SPXL and TNA and SOXL. Thanks for your assistance!

BarTune1
403 posts
msg #107657
Ignore BarTune1
modified
8/15/2012 11:03:40 PM

There are various threads a couple of years back on using the VIX divergences as a market entry timing indicator ... the variance vs. the 10 dma is derived from Connors. One of the better timing indicators for entering the SPY is using a BB setting of (9,2) and trading upon the spikes above/below - there is a thread on this.

There is another thread which considers the volitility indexes for oil, gold, the russell 2000 etc.

As regards the VIX and the VXX; the two are different animals and do not trade in tandem although many mistakenly think the two are the same. There is a pretty good free discussion about this on a video recently produced by Connors. See the video at the bottom of the web page here:

http://store.tradingmarkets.com/courses/etf/connors-research-volatility-trading-strategies-summit-b.html#vid

The video runs around 30 minutes - there is some excellent information in it if you intend upon trading the VXX or the XIV.

Consequently, I'm not sure it would be best to trade the VXX off a VIX reading. It may be best just to trade the short version as was originally posted.



BarTune1
403 posts
msg #107798
Ignore BarTune1
8/24/2012 3:52:08 PM

Nice ... the 97% system hits again with a 1 day trade garnering 4.49% as at 3:50 PM

novacane32000
273 posts
msg #107864
Ignore novacane32000
8/30/2012 8:40:50 PM

VXX RSI(2) closed above 90 today.

BarTune1
403 posts
msg #107865
Ignore BarTune1
8/30/2012 9:41:40 PM

yes, XIV triggered a buy at 13.48 at the close .... I will likely buy some tomorrow if i can get it on a further pullback

StockFetcher Forums · Filter Exchange · HELP FILTER 97% Success<< 1 2 3 4 5 ... 8 >>Post Follow-up

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2016 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus