The stock market is down and the VIX is up. The fear index is telling us something, but the issue is whether this is a signal that should be acted upon. The VIX index is volatile so any signal will have to be strong to offset the high volatility. The VIX history has gone through some long trends and the current increase has not broken the current long-term downtrend. Short-term spikes in the VIX are often quickly reversed and do not always represent a strong signal on asset allocation.
The use of the VIX as an asset allocation tool has had mixed results. There is no question that delevering and selling risky assets when the VIX reached high levels was a good trade. The late 2008 VIX signal was extremely valuable, but we are now almost six years from the crisis and the VIX has become a trickier signal to use.
First, if you are using any signal tool that looks back over time, you will be strongly biased by the 2008 spike. The mean and volatility both moved higher in the post-2008 period. A threshold model will not generate signals in this environment. Do you throw out the data and lose information or do you average in these extremes? It makes a big difference. Second, the form of the signal matters. There is a strong contemporaneous relationship between VIX spikes and stock declines but if you do not act immediately, you will have a very different pay-off function. It is the lag between VIX changes and returns that matters most if you want to have a useful signal.
However, the VIX is mean-reverting and the level has a lot of room to climb especially if sentiment further changes to the negative.
No comments:
Post a Comment