CBOE SKEW Index, Part 2

Read the first part of SKEW post here, part 3 here.

In the press release CBOE described SKEW index as "a benchmark measure of the perceived risk of extreme negative moves – often referred to as "tail risk" or a "black swan" event ". I guess they really needed a catchy moniker similar to "fear index" for VIX. However the main measure of an outlier is in volatility, not skew. SKEW index elimenates skew dependence on volatility, and there is almost no correlation between SKEW and VIX. What SKEW measures is the degree of assymetry and risk aversion that is not already reflected in VIX. By the very construction SKEW index is not redundant to the VIX.

If we apply SKEW in combination with VIX to market analysis (for example using it as technical indicator) we can see some interesting patterns. I separate SPX returns into 4 categories and calculate annualized returns (I assume everyone knows the importance of using geometric average returns for evaluation of long-term strategies)

Regime	Arithmetic average	Geometric average	Sharpe ratio
Low VIX Low SKEW	1.9%	1.3%	0.18
High VIX Low SKEW	12.3%	9.5%	0.49
Low VIX High SKEW	11.3%	11.3%	1.07
High VIX High SKEW	5.2%	2.7%	0.23

Right now we're in the Low VIX High SKEW regime, which can be considered a bullish signal. This is just the basic way of using SKEW index, more complicated indicators can be developed. For example Bollinger bands indicators seem to work reasonable well. In the next skew post I'll examine what SKEW means for option traders.