Contributions published at Financial Engineering (Markus Leippold)

null

null

null

Existing shrinkage techniques struggle to model the covariance matrix of asset returns in the presence of multiple-asset classes. Therefore, we introduce a Blockbuster shrinkage estimator that clusters the covariance matrix accordingly. Besides the definition and derivation of a new asymptotically optimal linear shrinkage estimator we propose an adaptive Blockbuster algorithm that clusters the covariance matrix even if the (number of) asset classes are unknown and change over time. It displays superior all-around performance on historical data against a variety of state-of-the-art linear shrinkage competitors. Additionally, we find that for small and medium-sized investment universes the proposed estimator outperforms even recent nonlinear shrinkage techniques. Hence, this new estimator can be used to deliver more efficient portfolio selection and detection of anomalies in the cross-section of asset returns. Furthermore, due to the general structure of the proposed Blockbuster shrinkage estimator the application is not restricted to financial problems.

null

In dieser Bachelorarbeit wurde untersucht, ob sich die Low-Risk Anomalie auf den Swiss Market Index anwenden lasst. Um dies zu uberprufen, wurden die Unternehmen des Swiss Market Index im Zeitraum von 2008 bis 2018 anhand ihres idiosynkratischen Risikos jahrlich in sechs Portfolios eingeteilt. Die Portfolios wurden mittels eines multiplen linearen Regressionsmodells gegen das Fama- Fench-Dreifaktorenmodell regressiert. Zwar sind die drei Faktoren des Fama-French Modells hoch signikant, aber die entsprechenden Alphas der Portfolios sind nicht signikant. Somit wird die Hauptthese dieser Bachelorarbeit, ob die Aktientitel des Swiss Market Index mit einem hoheren Risiko eine hohere erwartete Rendite haben, weder bestatigt noch verworfen.

null

null

null

We reexamine the existence of an upward trend in the idiosyncratic volatility (IV) of US-listed stocks between 1962 and 2000 documented in \textcite{CLMX2001}. Following the same methodology as in their paper, we confirm the upward trend and the subsequent reversal, as reported by \textcite{BHZ2012}. However, taking a closer look, we find that this result is influenced by microstructure biases in realized variances and correlations based on daily returns. Correcting these biases, we find that, depending on whether we use equal- or value-weighted IV and whether we include the pre-NASDAQ period, between 17\% and 62\% of the trend is removed. Also, the subsequent reversal of IV is almost entirely removed once we correct for these biases. We also highlight the mechanical dependence of the IV measure on industry concentration. From 1962 to 2000, industry concentration also contributes to a higher IV. However, its effect remains quantitatively small.

null

We estimate a flexible affine model using an unbalanced panel containing S\&P 500 and VIX index returns and option prices, and analyze the contribution of VIX options to the model's in- and out-of-sample performance. We find that they contain valuable information on the risk-neutral conditional distributions of volatility at different time horizons, which is not spanned by the S\&P 500 market. This information allows enhanced estimation of the variance risk premium. We gain new insights on the term structure of the variance risk premium, present a trading strategy exploiting these insights and show how to improve S&P 500 return forecasts.

null

null

null

The concept of second-order risk operationalizes the estimation risk in portfolio construction induced by model uncertainty. We study its contribution to the realized volatility of recently developed risk parity strategies. For each strategy, we derive closed-form solutions for the second-order risk, subsequently illustrated in empirical analysis based on real market data. The results suggest a relation between the contribution of second-order risk and the sensitivity of a portfolio to single eigenvectors of the covariance matrix of assets' returns. Among the strategies considered, we find the principal risk parity strategy, that invests equally in each eigenvector underlying the variance-covariance matrix, to be immune to second-order risk. For the other strategies, second-order risk can be partially mitigated by means of statistical methods.

null

null