Regular readers of this column know that I pay a lot of attention to the VIX, trying to unearth clues to the direction of the broad market. If you've been following the Market Monitor over the past few weeks, you've been treated to a regular series of comments from Linda Piazza, Kent Barton and Steven Price. Some of Linda's recent comments got us to talking about the VIX and to what degree technical indicators and chart patterns can be reliably used on the VIX.
The primary reason why this is even a subject for debate is that the VIX is not a tradable vehicle. It's fluctuations are actually derived from action in the options market, which is really derived from fluctuations in the price of the S&P 100 index (OEX.X). If looking for more details than any rational human should want to know about the creation of the VIX, feel free to check out the links below, where I went into great detail about how this index is calculated.
But I don't want to talk about any of the basis of the VIX today. While interesting to those with a mathematical bent, the important question for traders to answer is "How do I use the VIX?" That is the question about which Linda and I have debating recently. Before I delve into this important topic though, let's take a short detour into an interesting observation I came to. Linda and I both come from a fairly solid mathematical background, with a bent towards the realm of Physics. I've reflected in the past that oscillators (in one way of thinking of them) are a measure of the rate of change of price action. Relating this to the world of physics, there may be an interesting parallel. Speed or velocity of an object is a measure of the rate of change of position. Or for those of you with an advanced math background, velocity is the derivative of position. Remember that? Well, how about this? Acceleration (the rate of change of velocity) is the derivative of velocity. Simple calculus gives us a mathematical relationship between position, speed and acceleration in the realm of physics.
So what does any of this have to do with the VIX? Well the option market is part of the "Derivatives" market, and the VIX is really a derivative of the options market. So my inquisitive mind asked the question, "Is there a mathematical relationship between the price of the OEX, OEX options and the VIX that is roughly equivalent to the relationship between position, velocity and acceleration in the Physics world?" It may be a silly question, and probably uninteresting to many of my readers, but these are the strange kinds of questions that tend to occur to me late at night when I should be staring at the backs of my eyelids. I don't want to dwell on the issue too much here today, but if any of you have any bright ideas or thoughts on the subject, I'd certainly be interested to hear them. Hey, I'll bet Linda would find it interesting too!
Alright, let's get back to the subject at hand, that of how do we use the VIX. The basic question that Linda and I have been debating is "Since the VIX is really derived from the price action in other vehicles (OEX and OEX options) how reliably can we apply the study of chart patterns, oscillators and the use of moving averages?" I think Linda fired the first volley in this debate a few months ago, when she started following the action of the VIX as it related to its 200-dma. That didn't really get my attention, because at the time, I didn't see the 200-dma as providing much meaningful information with respect to the support/resistance being observed in the VIX. But it was enough to get my attention and I started looking at the VIX and its 200-dma.
Then a funny thing happened. I noticed that the 200-dma seemed to act as a reference point from which the rubber band (VIX) tended to get stretched, with particularly interesting results when it was stretched to the downside. I made note of this in the LEAPS column commentary back on January 26th, and I've copied the pertinent part of my observations here.
The one thing I do know is that throughout the time period from 1998 through late 2002, the 200-dma of the VIX tended to oscillate around the 25-26 area, sometimes rising as high as 29-30, and sometimes dropping as low as 23.50. But throughout that period of time, it remained confined to this range. But something interesting happened in the latter half of the year we just closed out. The VIX soared above 30 in early July and it took nearly 4-1/2 months to come back under that level. In the intervening span of time, the 200-dma rose through the 30 level for the first time ever and continues to rise, now at 33.01.
I looked at all the historical data for the VIX, and found that it has never been more than 8 points below its 200-dma. That occurred back in July of 2001, with the VIX at 20 and the 200-dma at 28. Doing a quick calculation from that difference, I come up with a minimum possible VIX (with the 200-dma at 33) of 25. Allowing for the increase in the baseline with the 200-dma now at 33 (a percentage increase of just under 18%, I come up with a maximum possible deviation from the 200-dma of 9.4. That means that with the 200-dma at 33, the absolute minimum VIX we could see would be around 23.5.
See the logic there? Throughout the past 5 years, the position of the VIX's 200-dma has helped to define a floor for the VIX itself. Up until recently the 200-dma has been confined to a pretty steady range between 23-29, and that has helped to keep the VIX in the 20-30 range most of the time. But with the significant increase in volatility in the overall market in recent months, the 200-dma is persistently rising (now just below 34) and the net result is that the floor for the VIX is also rising. So I think the 200-dma of the VIX is an important indicator, but not necessarily as an absolute support or resistance level.
Linda made another interesting observation last month, when she pointed out the potential for a double bottom formation near the 26 level. Sure enough, that seems to have been a very prescient observation, as the VIX soared through the 36 level (the peak between the two valleys of that formation) in late January, and since then has used that level as support.
Another interesting observation as it pertains to that double bottom formation is the fact that Stochastics gave pretty solid "Buy" signals as it moved up from the 26 area, both in late November and then again in the middle of January. However, note that when the Stochastics rolled over from overbought in both early December and then in late January, those "Sell" signals were much less reliable. So perhaps the Stochastics (and other oscillators) can provide solid "trading" signals, but only in the direction of the primary trend.
Finally, the descending trendline on the VIX from the July and October highs appears to be an important level, as it has continued to cap the upwards move in the VIX, even with the broad markets continuing to break support levels, getting closer and closer to the October lows. My expectation is that the VIX will eventually break above that trendline, and when it does, should come with some force. Remember that I've occasionally talked about the PnF chart on the VIX. It is telling us something interesting as well, as a print at 41 would be another PnF Buy signal. we've been close a couple of times in the recent past, but I think it is interesting that it hasn't been reached just yet. Another important clue? Time will tell!
So what do we know now that we didn't know 1400 words ago? I'm not sure we can draw any firm conclusions, but it certainly appears that there is merit to applying our tried and true knowledge about technical analysis to the VIX. Any observations we draw there should be used in the context of other observations we make directly on price action in the OEX, but I would view chart patterns and indicators on the VIX as just one more tool that we can use in our pursuit of understanding what the market is trying to tell us. Today's discussion isn't so much about providing a definitive answer as it is about opening a new door in the process of discovery. I hope you found this as provocative as I did. If nothing else, it sure is interesting!
Questions and observations are certainly welcome!