Session 3: Portfolio Theory

Chair:  Stepan Mazur

Christoph Frey: Shrinkage Estimation in Risk Parity Portfolios

Abstract: We investigate the impact of shrinkage estimation techniques for the moments of asset returns on risk-parity portfolios. In contrast to mean-variance port- folios, the risk contributions of individual assets in risk-parity portfolios are fixed a priori. This additional restriction is commonly found to stabilize empirical portfolio weights in time. We show that the marginal risk-budget for each portfolio asset indeed serves as a natural shrinkage target and hence provide a new perspective on risk-parity portfolios. In an extensive empirical application, we compare and combine the various shrinkage strategies to popular risk-based approaches from the literature. We find that while using shrinkage estimators in risk-parity portfolios enhances out-of-sample performance based on various criteria, traditional covariance shrinkage estimators dominate all other strategies in high-dimensional settings.

Bruno P. C. Levy: Trend-Following strategies via Dynamic Momentum Learning

Abstract: Time series momentum strategies are widely applied in the quantitative financial industry and its academic research has grown rapidly since the work of Moskowitz, Ooi, and Pedersen (2012). However, trading signals are usually obtained via simple observation of past return measurements. In this article we study the benefits of incorporating dynamic econometric models to sequentially learn the time-varying im- portance of different look-back periods for individual assets. By the use of a dynamic binary classifier model, the investor is able to switch between time-varying or constant relations between past momentum and future returns, dynamically combining or selecting different momentum speeds during turning points, improving trading signals accuracy and portfolio performance. Using data from 56 future contracts we show that a mean-variance investor will be willing to pay a considerable management fee to switch from the traditional naive time series momentum strategy to the dynamic classifier approach.

Maziar Sahamkhadam: Socially responsible multiobjective optimal portfolios

Abstract: We extend the socially responsible multiobjective problem to (i) estimate optimal portfolios via reward/risk maximization, (ii) include dependence structure between asset returns using vine copulas, and (iii) incorporate enhanced indexation utilizing cumulative zero-order stochastic dominance. In an application of the MOP optimization to a sample of Eurostoxx 50 constituents, we show that the optimal MOPs provide investors with the flexibility of incorporating deferent objectives. However, there is a trade-off between reward (risk) measures. Although, including social responsibility results in lower portfolio return and economic performance, it reduces the portfolio risk. While the cumulative zero-order SD objective (in most cases) increases the portfolio return when included in socially responsible MOPs, it reduces the portfolio risk. The predictive models lead to MOPs with higher return and reward/risk ratios. In particular, the copula-based MOPs achieve less tail risk.