| StockFetcher Forums · General Discussion · VIX variance vs. 10 day moving average | << 1 2 3 4 5 >>Post Follow-up |
| mystiq 650 posts msg #78926 - Ignore mystiq |
9/10/2009 1:05:59 AM 9/9/2009 9:13:58 PM vix ma(10) +5% -5% on plot Kevin_in_GA modified 9/9/2009 9:35:14 PM Very interesting correlation between RSI(2) and the VIX moving above/below is MA(10). Plot it out and see - on SPY it is like a mirror image. When the VIX is below is MA -5%, the RSI(2) is at or above 90. Converse is true as well - RSI(2) at 10 or less when the VIX is above the MA +5%. BarTune1 9/9/2009 10:28:06 PM As I said, this is one of my #1 indicators ..... I added short positions today in EWT, MDR, LINTA & KSU. And I bought the VXX etf. I think it is particularly "cheap" here. I will slowly add more positions if the market continues to go up. My primary screening criteria for shorting is RSI(2) > 98, overbought stocks. Using Larry Connors trading rules, my exit is end of day when the stocks cross the 5 day moving average. I am batting around 90% winning trades following these rules. (-.-) this one goes into the LIBRARY ! Thx guys for the coding & insight. |
| KMorley 6 posts msg #79077 - Ignore KMorley modified |
9/11/2009 4:07:47 PM Very nice work! I hate to ask a stupid newbie question, but I want to make very sure I understand the plot and the concept. As of this afternoon, the plot shows the current VIX at 24.12 and well under the VIX MA(10) at 25.82. I calculate the variance as current VIX being 6.6% under the VIX MA(10). If I understand correctly, that's an indication that the market is overbought. Conversely, if the current VIX was substantially above the VIX MA(10), that would be an indication that the market is oversold. Do I have that right? Also, when I look at the plot it's about the size of a book of matches. Is there any way to view the plot larger or full-screen? Thanks! |
| BarTune1 441 posts msg #79090 - Ignore BarTune1 |
9/11/2009 6:35:16 PM You have the relationship right .... I dont focus on the plot myself ..... only the extent to where it is to its 10 dma. It is useful to guide you in timing index or equity trading. The 5% rule (according to Larry Connors) basically says to be careful buying anytime the VIX is 5% below its 10 dma because since 1995 the S&P 500 has lost money one a net basis 5 days following the time the VIX has been 5% below its 10 dma. Conversely, whenever the VIX has been 5% or more above its 10 dma, the S&P has achieved returns which are better than 2-1 compared to the average weekly returns or all weeks. All this provides is an edge in timing entries, both long and short. Personally, I take notice when the divergence is 5% or greater . If it goes more than 10% I get fairly aggressive. Of course, I must note that my trading style is short term reversion to the mean. At the other end of the spectrum, there are those who favor momentum trading. Its all personal preference and whatever works, go with it. Kevin's idea of monitoring the divergence via Bollinger Bands seems quite useful. |
| Kevin_in_GA 4,599 posts msg #79093 - Ignore Kevin_in_GA modified |
9/11/2009 8:06:53 PM BarTune: You will enjoy this. I ran an optimization on trading the ^VIX as you described (short the VIX when it goes above the upper BB, cover when it closes back below the MA, long when it closes below the lower band, sell when it closes above the MA). Took data on the VIX gong back two years, and allowed both the MA and SD to vary. Below is the most profitable combinations: Optimization: BB-OPT-TS (09/11/2009 19:46:02) Trading System: VIX-Based_System Iterations: 336 Time: 0:05 Symbols Tested: 1 (QP:vix) From [09/12/2007 09:30] to [09/11/2009 16:00] Average: 2857.8, Best: 14177 V#1 V#2 APGT [Net] %PROF TRADES 9 1.3 4.66% $14177.28 83% 61 9 1 4.44% $13506.16 81% 61 9 1.6 5.1% $13483.26 79% 53 9 1.2 4.47% $13374.85 81% 60 9 2 7.42% $13326.05 88% 36 9 1.4 4.68% $13291.12 82% 57 9 1.1 4.37% $13280.55 81% 61 9 1.8 6.01% $13196.33 81% 44 9 1.5 4.53% $12654.30 82% 56 6 1.5 3.74% $12307.06 75% 66 9 1.9 5.8% $11849.25 80% 41 8 1.7 4.76% $11637.07 81% 49 8 1.2 3.77% $11460.27 81% 61 8 1.8 4.89% $11457.63 80% 47 5 1.4 3.12% $11342.25 73% 73 7 1.6 4.19% $11277.90 77% 54 6 1.6 3.8% $11171.36 76% 59 5 1.3 2.97% $10963.03 74% 74 7 1.3 3.39% $10649.07 76% 63 8 1.9 5.06% $10600.39 80% 42 9 1.7 4.71% $10330.02 72% 44 7 1.5 3.54% $10250.18 75% 58 7 1.7 4.19% $10247.02 77% 49 8 1.1 3.35% $10198.54 78% 61 8 1.6 4.08% $10160.87 80% 50 6 1.2 3.03% $10123.79 73% 67 8 1 3.27% $10103.27 80% 62 7 1.4 3.24% $10029.41 75% 62 I have highlighted what I think is the best combo - highest average gain per trade, an overall accuracy of 88%, only 36 trades required in the last 2 years, and it is among the most profitable after taking commissions out. Consider using BB instead of a moving average envelope, with the MA set to 9 and the SD at 2.0. Easy settings to remember. Now just reverse trade SPY based on where the VIX goes, and presto - money. Kevin |
| BarTune1 441 posts msg #79100 - Ignore BarTune1 |
9/11/2009 9:33:14 PM Thanks Kevin, I appreciate your insight ..... I have not backtested any particular strategies as I'm not all that handy with the software yet .... working a day job and having a family doesn't allow too much extra time for the market .... I will be monitoring the VIX bollenger bands closely now .... I saved that one screen from Stockcharts.com that you posted. I actually used that site fairly extensively for charts before joining Stock Fetcher ...... If you had the discipline to trade only the SPY or VXX based on your bollenger band strategy .... I think that is all you would really need to do ..... Funny how some of the best systems are the simplist systems. James |
| welliott111 98 posts msg #79124 - Ignore welliott111 |
9/12/2009 3:24:22 PM @ BarTune1 Are these accurate for some of Connors CVR signals? You seem to be the guy to ask for Vix info. The rules for the CVR 1 are simple: For market buys, we are looking for the VIX to make a 5-day high and close under its open. For sells, we are looking for the VIX to make a 5-day low and close above its open. The rules for the CVR 3 are as follows: For Buys: 1. Today, the low of the VIX must be above its 10-day moving average. 2. Today, the VIX must close at least 10% above its 10-day moving average. 3. If rules 1 and 2 are met, buy the market on the close. 4. Exit (on the close) the day the VIX trades (intraday) below yesterday's 10-day moving average (reversion to the man). Or exit within two to four days. For Sells: 1. Today, the high of the VIX must be below its 10-day moving average. 2. Today, the VIX must close at least 10% below its 10-day moving average. 3. If rules 1 and 2 are met, sell on the close. 4. Exit (on the close) the day the VIX trades (intraday) above yesterday's 10-day moving average (reversion to the mean). Or exit within two to four days. |
| Kevin_in_GA 4,599 posts msg #79128 - Ignore Kevin_in_GA |
9/12/2009 5:18:39 PM If you had the discipline to trade only the SPY or VXX based on your bollenger band strategy .... I think that is all you would really need to do ..... +++++++++++++++++++++++++++ If I were to trade on this, I would trade SSO/SDS (or even better UPRO/SPXU). These trades are short term, so why not use the leveraged ETFs? |
| BarTune1 441 posts msg #79130 - Ignore BarTune1 |
9/12/2009 8:26:26 PM welliot, I don't trade the CVR strategy, although I am aware of it ..... I merely follow the VIX as a market sentiment indicator ..... I subscribe to Trading Markets Power Ratings as one of my primary stock screening resources. Consquently I often buy their stocks and ETFs ranked 9 and 10 and sell the 1s and 2s. I will buy the VXX if it starts a lower divergence from its MA is it is currently and have been successful with my trades. I have Connors books, Short Term Trading Strategies that Work and High Probability ETF Trading .... I would highly recommmend them to anyone who wants to short term trade - reversion to the mean. My VIX knowledge comes from one of the chapters in Short Term Trading Strategies That Work .... and from some acedemic papers written relating to the VIX and Portfolio Managment Theory ..... |
| Eman93 4,750 posts msg #79140 - Ignore Eman93 modified |
9/13/2009 11:00:03 AM SSO over VIX  |
| BarTune1 441 posts msg #79144 - Ignore BarTune1 |
9/13/2009 2:00:08 PM That looks pretty nice Eman, I should just forget the rest of the trading I do .... and trade the SPY leveraged Proshares exclusively when the VIX is significantly extented or overextended (i.e., like greater than 10% divergence) .... I would probably do as well or better than i am currently, keeping the exit at closing the trade at the end of the day when it crosses the 5 dma. I have been trading primarily Trading Markets Power Ratings stocks that are either oversold or overbought when the VIX diverges .... maybe I should just stick to the index etfs .... there would be definate advantages to that .... including portfolio risk management primarily .... As an aside, when I was reading about optimal portfolio management theory, most academic papers indicated that every long portfolio should consist of approximately 10-15% VIX futures as a solid hedge .... or as is the case now, you can play the VXX ..... Its pretty interesting, as had one hedged a long portfolio with the VXX (although it has only been available as a recent creation), when the market dropped in Sept Oct 2008, they would have seen their portfolio increase by what, say 80/15 or maybe 4 to 5 times on the VXX position as it jumped from around 15-20 to 80 at one point ..... probably would have offset all losses or more on long side positions .... in fact, you may have ended up with a profit on a long portfolio during the meltdown Most long portfolios would probably have lost nothing during last falls meltdown if hedged with 15% VXX ..... holdings .... Interesting theory .... Personally, I am not concerned with a long portfolio, as i am short as often as i am long, and normally for only a short period. As Kevin's Bollenger Band System is primarily a long strategy, it could possibly benefit from VXX hedging @ 10-15% of the total portfolio ..... I recognize the outstanding returns on his system and am amazed. I would suspect the only risk is a somewhat prolonged downtrending market ....... I guess the other risk would be a market crash .... and if all long positions were hedged with minor VXX component .... I realize it might hurt returns slightly (as all insurance costs money) but it would guard against any coming of the Black Swan ..... (i.e., say he had $100,000 in open positions under the system when the market dropped last fall and say $15,000 was comprised of the VXX) ...... with the VIX going up around 80 at one point ... the VXX insurance might have gone to $60,000 if it were bought when the VIX was 20 .....) .... |
| StockFetcher Forums · General Discussion · VIX variance vs. 10 day moving average | << 1 2 3 4 5 >>Post Follow-up |