Stock portfolios formation in S&P100, macroeconomic variables predictability role in returns and beta coefficients time-variability test: an empirical approach

Abstract

The S&P100 Stock Index is the core of the largest companies in the U.S. Market in terms of capitalization (blue chips), characterized by relatively strong stability in their stock price changes. However, due to the different economic sectors of activity of its companies, there are sectors with different volatility behavior. This phenomenon is noticed at companies in the Financial Sector of S&P100 (less stability), in line with those of Information-Technology (greater stability), that seems to appear two of the highest S&P100 Sector’s volatility returns. This study, initially classifies three different portfolio categories in relation to their structure for both of these S&P100 sectors, then forms nine different stock portfolios for each category, based on the annual prices of key fundamental stock indicators, re-adjusted each year. Descriptive Statistics of all portfolio monthly returns are presented and appear a controversial result regarding with the relation risk-return. On the one hand the purpose of this research is to examine the possible effect on portfolio returns formed due to American macroeconomic variables, on the other hand to look for a possible alteration in their individual sensitivity factors (betas) through time. This time variability in betas is attempted to be identified both visually with the graphical representation of the produced beta coefficients through a Rolling Regression process with a 60-month calendar window, as well as econometrically with the model estimation that linking beta coefficients with time for all portfolios formed. The results of this study are useful tools for both investment and academic level when considering the factors that influence portfolio returns on the S&P100 and the influence of time on them, particularly in moments of crises such as the one in 2009, combined with the level of risk tolerance even in shares that historically demonstrate limited price volatility.