Asset Allocation under Parameter Uncertainty
organized by
Alex Weissensteiner, Free University of Bozen-Bolzano, Italy
venue: Free University of Bolzano - Room F6 University Club
date: Dec 18, 2017
map: see below
Invited speakers
Nicole Branger - University of Muenster
Thomas Dangl - TU Wien
Lorenzo Garlappi - Sauder School of Business
Michael Hanke - University of Liechtenstein
Tim A. Kroencke - University of Basel
Claus Munk - Copenhagen Business School
Rolf Poulsen - University of Copenhagen
Sebastian Stöckl - University of Liechtenstein
Programm
Sunday, Dec 17 come together
19:00 meeting in the lobby of the Hotel Stiegl
19:30 dinner Römerkeller
Monday, Dec 18 scientific program
09:00 - 09:15 Welcome speech by Oswin Maurer (dean of the faculty of economics)
09:15 - 10:00 Parameter Uncertainty, Financial Turbulence and Aggregate Stock Return (S. Stöckl)
10:00 - 10:45 Idiosyncratic Volatility, its Expected Variation, and the Cross-Section of Stock Returns (N. Branger)
10:45 - 11:15 coffee break
11:15 - 12:00 The FOMC Risk Shift (T. Kroencke)
12:00 - 12:45 Risk-minimization in Electricity Markets: Fixed Price, Unknown Consumption (R. Poulsen)
12:45 - 14:15 lunch break
14:15 - 15:00 Comparing Large-Sample Maximum Sharpe Ratios and Incremental Variable Testing (M. Hanke)
15:00 - 15:45 Hedging Recessions (C. Munk)
15:45 - 16:15 coffee break
16:15 - 17:00 Long-Term Asset Allocation under Time-Varying Investment Opportunities: Optimal Portfolios with Parameter and Model Uncertainty (T. Dangl)
17:00 - 17:45 Dynamic Consumption and Portfolio Choice under Model and Parameter Uncertainty: Learning about Return Predictability (L. Garlappi)
17:45 - 18:00 discussion
19:30 dinner Vögele
Abstract of invited talks:
Idiosyncratic Volatility, its Expected Variation, and the Cross-Section of Stock Returns (SSRN)
Nicole Branger, Hendrik Hülsbusch and Frederik Middelhoff
We offer a novel perspective on the negative relation between idiosyncratic volatility (IVOL) and expected returns. We show that the IVOL puzzle is largely driven by a mean-reversion behavior of the stocks' volatilities, which is not captured by a simple historic measure of IVOL. In doing so, we make use of option implied information to extract the expected mean-reversion speed of IVOL in an almost model-free fashion. Together with the current level of IVOL this method allows us to identify stocks' expected IVOL innovations in a very general setting. Under the assumption of IVOL carrying a positive price of risk (Merton (1987)) we resolve the puzzle. In a horse race we show that the mean-reversion speed is superior to the most prominent competing explanations. All our findings are robust to different measures of IVOL and various stock characteristics.
Long-Term Asset Allocation under Time-Varying Investment Opportunities: Optimal Portfolios with Parameter and Model Uncertainty (SSRN)
Thomas Dangl and Alex Weissensteiner
We study the implications of predictability on the optimal asset allocation of ambiguity averse long-term investors. We analyze the term structure of the multivariate risk-return trade-off in a VAR model under full consideration of parameter uncertainty, and we decompose the predictive covariance along different sources of risk/uncertainty. We calibrate the model to real returns of US stocks, US long-term government bonds, cash, real-estate and gold using the term spread and the dividend-price ratio as additional predictive variables. While over short periods the model-implied conditional covariance structure of asset-class returns determines the optimal allocation, we find that over longer horizons the optimal asset allocation is significantly influenced by the covariance structure induced by estimation errors. As a consequence, the ambiguity averse long-term investor tilts her portfolio not simply toward the global minimum-variance portfolio but shrinks portfolio weights toward a seemingly inefficient portfolio which shows maximum robustness against estimation errors. Most interestingly, we find that even though time diversification of stock returns vanishes after consideration of estimation errors, real long-term bond returns are even more affected, making stocks an important asset class for the ambiguity averse long-term investor.
Dynamic Consumption and Portfolio Choice under Model and Parameter Uncertainty: Learning about Return Predictability
Lorenzo Garlappi and Georgios Skoulakis
In this paper we consider the lifetime consumption and portfolio choice of an investor who is uncertain about both the parameters driving asset returns and the model underlying the data generation process. The setting is motivated by the long-standing debate in the empirical asset pricing literature regarding the existence and strength of predictability in asset returns. If returns are predictable, portfolio allocations would crucially depend on the dynamics of the predictor as well as the investment horizon. However, even assuming that the investor is convinced about the existence of predictability, a further issue remains regarding the strength of such predictability. We study this issue by solving the dynamic consumption and portfolio choice problem for a long-lived investor who is uncertain both about the model generating the return data and the values of all of the parameters describing each model. We view these two layers of uncertainty as essential features of virtually any realistic consumption and portfolio choice problems. We show that the dynamics of the learning problem in this relatively simple setting (one risky-asset), is fully described by 8 state variables, rendering the solution numerically challenging. We explore the use of recent techniques in theory of Gaussian Processes to efficiently handle the high dimensionality of the problem.
Comparing Large-Sample Maximum Sharpe Ratios and Incremental Variable Testing (SSRN)
Michael Hanke and Spiridon Penev
Most existing results on the distribution of the maximum Sharpe ratio depend on the assumption of multivariate normal return distributions. We use recent results from the literature to provide an analytical representation of the distribution of the difference between two maximum Sharpe ratios for much less restrictive distributional assumptions, both with and without short sales. Knowing the distribution of the difference enables us to test ex ante whether or not the inclusion of additional variables leads to a significant improvement in the maximum Sharpe ratio. In addition, we characterize the optimal long-only solution and provide conditions for global optimality.
The FOMC Risk Shift (SSRN)
Tim A. Kroencke, Maik Schmeling and Andreas Schrimpf
This paper presents new evidence on channels through which monetary policy affects prices in equity and other asset markets. A large part of U.S. equity price moves around FOMC meetings can be attributed to shocks that are uncorrelated with yield changes but closely linked to changes in investors' risk appetite. These price effects are mirrored by investors' portfolio rebalancing decisions, manifesting themselves via sizeable shifts in fund flows between bonds and equities. All these effects are transitory and largely reversed after about one month. We find evidence that risk appetite shocks are related to changes in uncertainty triggered by FOMC meetings.
Hedging Recessions (SSRN)
Nicole Branger, Linda Sandris Larsen and Claus Munk
Traditional life-cycle models conclude that consumer-investors should be fully invested in stocks when young---in stark contrast to observed stock holdings---and then gradually replace stocks with bonds as retirement is approaching. We show that a careful modeling of business cycle and unemployment risks reduces the early-life stock holdings dramatically, in some situations even to zero in accordance with the wide-spread non-participation in the stock market. The reduction is driven by the relatively high unemployment risk of young adults, the negative influence of unemployment on future salaries, and the business cycle variations in stock prices and unemployment risk that cause human capital to be riskier and more stock-like.
Risk-minimization in Electricity Markets: Fixed Price, Unknown Consumption (Paper)
Martin Jönsson, Rolf Poulsen, Rune Ramsdal Ernstsen, and Anders Skajaa
We analyze risk management of fixed price, unspecified consumption contracts in energy markets. We model the joint dynamics of the spot-price and the consumption of electricity, study expected loss minimization for different loss measures, and derive optimal static hedge strategies based on forward contracts. The strategies are implemented empirically and compared to a benchmark strategy widely used by the industry. On 2012--2014 Nordic market data, the suggested hedges significantly outperform the benchmark: The realized cumulative profit-and-losses are greater for almost every single one-month period and the hourly realized pay-offs result in an approximate 65% out-performance probability. Hedges based on asymmetric loss measures yield markedly higher reward-to-risk ratios than the benchmark, which can be exploited to release a premium from the contract in the financially significant order of 1.5% of the fixed price.
Parameter Uncertainty, Financial Turbulence and Aggregate Stock Return (SSRN)
Sebastian Stöckl
In this paper, we develop a novel, intuitive and objective measure of time-varying parameter uncertainty (PU) based on a simple statistical test. Investors who are averse to PU will react to elevated levels of PU by withdrawing from the market and causing prices to fall, a behavior that is well described by the model of portfolio selection with PU of Garlappi et al. (2007). We show that when combined with our measure, this model outperforms all other tested variables in predicting the equity premium, including the strongest known predictor to date. Additionally, it is the only predictor that fulfills all criteria generally expected from a stable predictor of the equity premium. All our results are statistically and economically significant and robust to a large variety of different specifications.
Participation is free of charge - but please register by sending an email to: alex.weissensteiner@unibz.it
Arrival via train: both the university and the hotel are within walking distance