eThesis - Kauppakorkeakoulun sähköiset gradut

» hse_ethesis_14074.pdf   koko: 2 MB (1515029)

This thesis examines the explanatory power of equity index options and volatility indexes on short-term equity index returns. The work relates to literature that employs options to show what risks are compensated in financial assets and how to assess the riskiness of investments in general. The contribution of this thesis is to provide a comprehensive assessment of the main sources of predictive power by employing a wide range of option-implicit predictors. Moreover, global evidence is provided. The predictive information is based on risk-neutral nonparametric probability distributions of equity index returns, volatility indexes, and realized returns. The main focus is in using information available before the return periods to explain subsequent returns.

The considered predictive variables comprise of option-implied volatility, option-implied skewness, volatility risk premium, skew risk premium, and two tail risk measures. This thesis answers whether the variables are consistently behaving and statistically significant predictors across indexes. The involved indexes are S&P 500 (the U.S.), FTSE 100 (the U.K.), and DAX 30 (Germany). In addition, information on S&P 500 is used to explain returns of ten equity indexes globally, the remaining seven being Euro STOXX 50 (Europe), Nasdaq OMX Helsinki (Finland), Hang Seng (Hong Kong), Nikkei 225 (Japan), MXIPC35 (Mexico), MERVAL (Argentina), and S&P/ASX 200 (Australia). The sample ranges from February 2006 to December 2014, and the option data includes options with expiries from May 2006 to December 2014. Predictive regressions are run for one-week, one-month, two-month, and three-month returns.

The results show that tail risk and volatility risk premium are the main sources of predictive power, and that the variables are robust to the inclusion of other option-implicit variables and market valuation, interest rate, and dividend based alternative explanatory variables. The results hold for FTSE 100 and DAX 30 only if the information in S&P 500 options is used to explain returns. This implies that risks and premiums are global, and that the information is best reflected in S&P 500 options. Applying the information in S&P 500 options on ten indexes globally further confirms this finding. A one standard deviation increase in the ex ante volatility risk premium of S&P 500 on average leads to a 2.2% to 4.1% increase in three-month logarithmic returns globally. A one standard deviation increase in the tail risk measure leads on average to a 3.2% to 7.8% increase. Skew risk premium seems to contain the same information as the volatility risk premium. The tail risk measure and the volatility risk premium also likely contain similar predictive information, as the inclusion of the tail risk measure at most leads to a 1% increase in explained variation for the ten equity indexes. The predictive power in the variables comes mainly from the time-varying risk of large price movements and risk aversion.