Forecasting European Stock Returns with Options Implied Volatility
Chatziparaskevas, Georgios
Pavlopoulos, Georgios
This dissertation was written as part of the MSc in Banking and Finance at the International Hellenic University. The dissertation topic is “Forecasting European stock returns with options implied volatility”. The main issue under scrutiny is the estimation of the extentwhich stock market returns can be predicted from a volatility measure based on implied volatility. The analysis focuseson the major economies of the European region. Major stock market and volatility indices of the region are used. More specifically the implied volatility is obtained fromthe benchmark volatility indices VDAX, VCAC, VFTSE, which are indicative of the three most important European markets (Germany, France, and United Kingdom), and VSTOXX, which is representative of the whole European Region. Research is conducted on whether the variance premium estimated by the volatility indices can be used as indicator of the behavior of the respective stock market indices (DAX, CAC, FTSE, STOXX). The relation between each volatility index of the European region and the underlyingstock market one is going to be investigated. In order to achieve that, the GARCH model is used to estimate the conditional variance of the stock market index for a given period. Afterwards the variance premium is calculated from the implied volatility index and the conditional variance. Having estimated the variance premium an effort to model its relation with the stock returnsby means of linear regression is made. If such a relation can be defined,evidence exists that there is an implied relation between the volatility index and the stock returns. The purpose of the current thesis is to investigate the extent of this relation and whether the volatility index translated to variance risk premium is a trustworthy indicator of future stock returns. The relation between the variance premium and stock returns is highly dependable on the estimation of the conditional variance as the latter is one of its components.A great part of the thesis is dedicated to the estimation of the conditional variance. After amodel investigation procedure is applied, it is found that GARCH is suitable and eventually is used in order to obtain the conditional variance.
