The CBOE S&P 500 PutWrite Index and a strategy

Pub: 17.09.15

In General.

A long write up leading up to the use of the PutWrite index was on Philosophical Economics. The PutWrite index is not widely used in academia, but it does seem like an interesting index for practical use.

There's a lot of text in the original post for what in the end is a fairly straightforward timing strategy of being long in the S&P 500 but switching to the PutWrite index when the VIX volatility index dips below 20. The indices and the strategy returns end up looking as follows: The data is from the end of February 1990 to August 2015 and the S&P 500 is the Total Return index. You can compare these values to the ones at the original post and see that they are similar.

Here I have also plotted the VIX index for your convenience: You notice that there is a quite a bit of variation and the average over the whole sample hovers around 20.

Here is a table with summary statistics for the indices and the strategy for comparison purposes. Taken together, the PutWrite index and strategy do seem like nice things to test out in your portfolio. However, the problems with the strategy become apparent with a year by year analysis: The strategy has only beat the S&P 500 8 years out 25, or 32 % of the time. You might have to wait a long time before you get as lucky again. The original post outlines various reasons why you would want to specifically do it now, however, naively speaking it is unclear whether this is a good time for the strategy or not.

Can we do better?

Of course we can. Let's assume a model where the one month lagged level of the VIX explains both the S&P 500 index and the PutWrite Index. Then, if we estimate the slope and intercept of the linear model with, for example, twelve months of data, then we can compare which forecast has a higher return and invest in that one. What does it look like? And if we do that we can get a hit rate at the annual level of 50 %. The strategy beat the S&P 500 12 years out of 24. A linear model allowed us to have a dynamically varying strategy, instead of just a naive fixed VIX switching level at 20. 