Theo Berger

On portfolio optimization: Forecasting asset covariances and variances based on multi-scale risk models

Purpose - The purpose of this paper is to show how investors can incorporate the multi-scale nature of asset and factor returns into their portfolio decisions and to evaluate the out-of-sample performance of such strategies. Design/methodology/approach - The authors decompose daily return series of common risk factors and of all stocks listed in the Dow Jones Industrial Index (DJI) from 2000 to 2015 into different time scales to separate short-term noise from long-run trends. Then, the authors apply various (multi-scale) factor models to determine variance-covariance matrices which are used for minimum variance portfolio selection. Finally, the portfolios are evaluated by their out-of-sample performance. Findings - The authors find that portfolios which are constructed on variance-covariance matrices stemming from multi-scale factor models outperform portfolio allocations which do not take the multi-scale nature of asset and factor returns into account. Practical implications - The results of this paper provide evidence that accounting for the multi-scale nature of return distributions in portfolio decisions might be a promising approach from a portfolio performance perspective. Originality/value - The authors demonstrate how investors can incorporate the multi-scale nature of returns into their portfolio decisions by applying wavelet filter techniques.

Oil price and FX-rates dependency

Oil prices and exchange rates against the dollar have both experienced long swings over the recent decade. Regardless of the great amount of research, some issues are still open to debate. In this vein, this paper focuses on the evolution of the relationship between oil prices and dollar exchange rates of 12 oil exporting and oil importing countries based on a dynamic copula approach. We use daily data for two 5-year periods between 2003 and 2013, taking the collapse of Lehman Brothers as the dividing point. Our results have four main implications: first, the intensity of relationship between oil prices and FX-rates has increased over time even if the peak of the financial crisis is included. Second, the increased tail dependency shows that extreme events are likelier to occur simultaneously for both series. Third, the dependency has become more dynamic after the financial crisis and is therefore better characterized by time-varying copulas. Finally, currencies of oil importers and oil exporters display a different dependency structure against the US dollar in the case of rising oil prices with the latter appreciating and the former depreciating.

Multivariate dependence and portfolio optimization algorithms under illiquid market scenarios

We propose a model for optimizing structured portfolios with liquidity-adjusted Value-at-Risk (LVaR) constraints, whereby linear correlations between assets are replaced by the multivariate nonlinear dependence structure based on Dynamic conditional correlation t-copula modeling. Our portfolio optimization algorithm minimizes the LVaR function under adverse market circumstances and multiple operational and financial constraints. When considering a diversified portfolio of international stock and commodity market indices under multiple realistic portfolio optimization scenarios, the obtained results consistently show the superiority of our approach, relative to other competing portfolio strategies including the minimum-variance, risk-parity and equally weighted portfolio allocations.

Does Gold Act as a Hedge or a Safe Haven for Stocks

This study deals with the issue whether gold actually exhibits the function of a hedge or a safe haven as often referred to in the media and academia. In order to test the Baur and Lucey [2010] hypotheses, we contribute to the existing literature by the augmentation of their model to a smooth transition regression (STR) using an exponential transition function which splits the regression model into two extreme regimes. One accounts for periods in which stock returns are on average and therefore allows to test whether gold acts as a hedge for stocks, the other one accounts for periods characterized by extreme market conditions where the volatility of the stock returns is high. The latter state enables us to test whether gold can be regarded as a safe haven for stocks. The study includes a broad set of 18 individual markets as well as five regional indices and covers a sample period running from January 1970 to March 2012 on a monthly frequency. Overall, our fi ndings show that gold serves as a hedge and a safe haven. However, this ability seems to be market-specific. In addition, by applying a portfolio analysis we also show that our findings are useful for investors.

Copulas and Portfolio Strategies: An Applied Risk Management Perspective

Modeling multivariate return distributions via copula functions is a common approach in financial risk management. However, evidence of the impact of choosing a particular copula function on different portfolio compositions is still lacking, as empirical studies typically analyze only a single portfolio strategy (eg, equally weighted portfolios) at a time. We evaluate copula models for three different portfolio strategies in a twenty-asset/daily return framework with respect to the accuracy of value-at-risk forecasts. From this risk management perspective, (dynamic) t copulas turn out to be a sensible choice for different portfolio strategies, especially when the trading strategy does not exclude highly volatile assets.

The macroeconomic role of currency reserve accumulation in emerging markets—The Asian experience

The impact of currency reserve accumulation is controversially discussed since reserve accumulation potentially destabilises the international financial system and causes crises due to higher systemic risk. The main aim of this paper is to put the macroeconomic role of currency reserve accumulation for four Asian economies under closer scrutiny. The key question is whether accumulating currency reserves is beneficial from a long†run perspective. Based on a vector error correction approach, we start by analysing long†run steady†state relationships between currency reserves, exchange rates against the US dollar, real GDP and interest rates. Our findings show that cumulated currency reserve shocks significantly affect real GDP. A likely explanation for our finding is that accumulation of reserves has supported growth through providing liquidity and supporting the development of the financial sector for the economies under observation.

Wavelet Decomposition and Applied Portfolio Management

In this paper, we decompose financial return series into their time and frequency domains in order to separate short-term noise from long-term trends. First, we investigate the dependence between US stocks at different time scales before and after the outbreak of financial crisis. Second, we set up a novel analysis and introduce the application of decomposed return series to a portfolio management setup. We then model portfolios that minimize the volatility of each particular time scale. As a result, portfolio compositions that minimize the short-run volatility of the first scales represent a promising choice, since they slightly outperform portfolio compositions that minimize the variance of the unfiltered return series.

Value-at-Risk Forecasts Based on Decomposed Return Series: The Short Run Matters

We apply wavelet decomposition to decompose financial return series into a time frequency domain and assess the relevant frequencies for adequate daily Value-at-Risk (VaR) forecasts. Our results indicate that the frequencies that describe the short-run information of the underlying time series comprise the necessary information for daily VaR forecasts.

Financial Crisis, VaR Forecasts and the Performance of Time Varying EVT-Copulas

We investigate Value-at-Risk (VaR) estimates based on extreme value theory (EVT) models combined with time varying parametric copulas against competing parametric approaches accounting for dynamic conditional correlations feasible to higher order portfolios. Tails of the return distributions are modeled via Generalized Pareto Distribution (GPD) applied to GARCH filtered residuals to capture excess returns, linked via constant and time varying copulas. Drawing on this EVT-GARCH-Copula, we evaluate portfolios consisting of German Stocks, market indices and FX-rates. However, the empirical results support the dynamic EVT-GARCH-Copula approach, as 99 % VaR forecasts clearly outperform parametric estimates stemming from competing dependency approaches.

Misspecified Dependency Modelling: What Does It Mean for Risk Measurement?

Forecasting portfolio risk requires both, estimation of marginal return distributions for individual assets and dependence structure of returns as well. Due to the fact, that the marginal return distribution represents the main impact factor on portfolio volatility, the impact of dependency modeling which is required for instance in the field of Credit Pricing, Portfolio Sensitivity Analysis or Correlation Trading is rarely investigated that far. In this paper, we explicitly focus on the impact of decoupled dependency modeling in the context of risk measurement. We do so, by setting up an extensive simulation analysis which enables us to analyze competing copula approaches (Clayton, Frank, Gauss, Gumbel and t copula) under the assumption that the “true” marginal distribution is known. By simulating return series with different realistic dependency schemes accounting for time varying dependency as well as tail dependence, we show that the choice of copula becomes crucial for VaR, especially in volatile dependency schemes. Albeit the Gauss copula approach does neither account for time variance nor for tail dependence, it represents a solid tool throughout all investigated dependency schemes.