Master's Thesis

In his Master's thesis, Nicholas Matthyssens investigated whether the stock return dispersion (the standard deviation of the returns of a group of equity portfolios) could provide useful information about the future equity premium. For this purpose, he used 25 European portfolios.

Nicholas: "To investigate whether the stock return dispersion was a good predictor, I performed univariate, bivariate, and multivariate long-run predictive regressions using the Ordinary Least Squares method. I was taught these quantitative research methods in my Master’s in Financial Business Engineering."

The regressions within the sample taken showed a significant association for the relationship Nicholas wanted to test. "But when the sample was expanded to include data from a longer time period, there was no significant relationship to be found. When I ran regressions on data outside the initial sample, the stock return dispersion only proved to be a reliable predictor for a period of one month, but not for the evolution over one or more years."

Finally, Nicholas looked at whether stock return dispersion has any economic value as a predictor. "Simply put, this means that I examined the average investor's willingness to pay for the use of the prediction model based on the 'stock return dispersion'. According to the theory, an investor is willing to do so if he has sufficient certainty that the return he achieves in that case is greater than the 'cost' of using the prediction model, and the return is also significantly more accurate compared to another valuation strategy." Nicholas' calculations showed that the 'stock return dispersion' is indeed also economically significant as a predictor, for the 25 European equity portfolios examined.