The performance of idiosyncratic volatility portfolios in the S&P 500
Stålnacke, Jarkko (2016)
Stålnacke, Jarkko
2016
Kuvaus
Opinnäytetyö kokotekstinä PDF-muodossa.
Tiivistelmä
The aim of this thesis is to test whether portfolios of S&P 500 stocks, sorted on idiosyncratic volatility, yield different future returns. According to traditional finance theory, higher undiversified risk is compensated with higher expected return. This suggests that portfolios containing high idiosyncratic volatility stocks provide higher future returns. Decile portfolios are formed based on one month’s lagged idiosyncratic volatility obtained from each stock’s historical return data. Idiosyncratic volatility is estimated as the standard deviation of the Fama-French (1993) three-factor model’s residual and the portfolios are rebalanced on a monthly basis. In addition, 15 beta-idiosyncratic volatility portfolios are formed to provide further evidence.
The time series results are analyzed within the framework of the Fama-French (1993) three-factor model. The interest of this thesis is to analyze the differences between portfolios’ risk-adjusted excess returns, i.e. alphas. Moreover, the portfolios’ holding-period performance is reported in the end of this thesis. The sample includes the daily returns of 280 stocks from the S&P 500, covering a time period from 1990 to 2013.
The findings of this thesis provide evidence that high idiosyncratic volatility portfolios generate positive and highly statistically significant risk-adjusted excess returns. That is, the portfolio alphas are positively correlated with idiosyncratic volatility in the sample. As the low idiosyncratic volatility portfolios did not show significant alphas, the zero-investment strategies (IVOL10-IVOL1 and IVOL9-IVOL1) had significant daily alphas of 0.06% and 0.03%, respectively. Despite the different risk-factors, the results were consistent throughout the paper. Furthermore, high idiosyncratic volatility portfolios provided best risk-return trade-off. However, during recessionary months, the risk-adjusted excess returns were insignificant in the sample.
The time series results are analyzed within the framework of the Fama-French (1993) three-factor model. The interest of this thesis is to analyze the differences between portfolios’ risk-adjusted excess returns, i.e. alphas. Moreover, the portfolios’ holding-period performance is reported in the end of this thesis. The sample includes the daily returns of 280 stocks from the S&P 500, covering a time period from 1990 to 2013.
The findings of this thesis provide evidence that high idiosyncratic volatility portfolios generate positive and highly statistically significant risk-adjusted excess returns. That is, the portfolio alphas are positively correlated with idiosyncratic volatility in the sample. As the low idiosyncratic volatility portfolios did not show significant alphas, the zero-investment strategies (IVOL10-IVOL1 and IVOL9-IVOL1) had significant daily alphas of 0.06% and 0.03%, respectively. Despite the different risk-factors, the results were consistent throughout the paper. Furthermore, high idiosyncratic volatility portfolios provided best risk-return trade-off. However, during recessionary months, the risk-adjusted excess returns were insignificant in the sample.