- 2001: Volume 10, No. 1 Volatility Index and Stock Market Bottoms?

Option volatility has long been discussed as a predictor of future market movements. Larry McMillan discussed this phenomeon in his article Option Volatility: How Good a Predictor (see CRB Trader September/October 2000).

We decided to examine the relationship between stock market volatility and the performance of the market itself.

Question
Do spikes in stock market volatility signal bottoms in the stock market?

Definitions
Before we could hand that question to the computer, a few terms needed to be defined.

Stock Market—We defined the stock market as the daily prices of the S&P 500 future.

Volatility—We defined volatility as measured by the Volatility Index. The VIX is calculated real-time and is equal to the average of the implied volatility of eight OEX options with about 30 days to expiration.

Volatility Spike—Techies have found when the VIX rises above 35 percent, it is a sign of a market bottom. How useful was that threshold years ago, when volatility averaged 10 percent and it never rose above 35? To standardize this measure we defined a volatility spike as the VIX rising more than 15 percent above its average level of the past 20 days, e.g., if the VIX averaged 20 percent the past 20 days and its most recent closing level was greater than 23 percent, volatility would be spiking.

Market Bottom—The stock market has marked a bottom when the price change over the following 10 days is positive.

Trading Strategy
The day the VIX spikes, the strategy gets long the S&P 500 future. The long position is then held 10 trading days. The position is then liquidated on the close of the tenth day.

Only one position can be held during the 10-day holding period, thus, additional volatility spikes while the strategy is long, are ignored.

Test and Results
We tested this strategy using closing S&P futures prices and VIX volatility levels from 1994 through 2000. In our simulation one S&P contract was traded, and no slippage or transaction costs were charged. (See Table 1.)

Table 1

The Numbers
Average Profit Per Trade	$3,150
% of Winners # of Winners # of Losers	73.3% 33 12
Avg. Win/Avg. Loss Avg. Winner Avg. Loser	1.06 $6,552 $6,203
Largest Winner Largest Loser	$23,757 $15,600
Most Losers in a Row Largest Drawdown Largest Loser	4 $28,072 $15,600
Net Profit	$141,772

The VIX spiked 45 occasions during the test period. Thirty-three times the stock market rose over the following 10-day period. Only 12 times did the stock market trade lower and the spike in volatility did not mark a bottom. (See Figure 1 for examples.)

Figure 1:

Comment:
Given the strategy holds its positions 10 days no matter what, it is not surprising the size of the average winning and losing trades are about the same. With a fixed holding period, a strategy has to be right often in order to generate a profit. This one is and its profitability supports the notion a spike in volatility often (73 percent of the time in this test) marks a market bottom.

Alex Saitta is a vice president and Yuxin Li is a research assistant in the New York office of Salomon Smith Barney. Alex can be reached at alex.saitta@ssmb.com