Contributions published at Quantitative Business Administration

null

null

null

null

DeMiguel et al. [DeMiguel V, Garlappi L, Uppal R (2009) Optimal versus naïve diversification: How inefficient is the 1/N portfolio strategy? Rev. Financial Stud. 22(5):1915–1953] showed that in the stock market, it is difficult for an optimized portfolio constructed using mean-variance analysis to outperform a simple, equally weighted portfolio because of estimation error. In this paper, we demonstrate that portfolio optimization can be made to work in currency markets. The key difference between the two settings is that in currency markets interest rates provide a predictor of future returns that is free of estimation error, which permits the application of mean-variance analysis. We show that over the last 26 years, a mean-variance efficient portfolio constructed in this fashion has a Sharpe ratio of 0.91, versus only 0.15 for the equally weighted portfolio. We also consider the practical implementation of this strategy.

Most standard asset-pricing models assume that all shocks to consumption are permanent. We relax this assumption and allow also for non-permanent shocks. In our specification, the long-run mean of consumption growth is constant; consumption levels are subject to short-run deviations from their long-run trend. The implications of our model are dramatically different from those obtained in the prior literature. A canonical and parsimonious asset pricing model with CRRA preferences and non-permanent shocks can reproduce the equity premium, high return volatility and return predictability with a coefficient of relative risk aversion below ten. This finding suggests that non-permanent shocks can play an important role in explaining asset pricing puzzles.

null

null

null

null

null

null

null

Asset management companies are constantly searching for portfolio optimization models that are on the one hand clear and intuitive and on the other provide high and reliable returns. This paper presents a modifed version of the well-known Black-Litterman portfolio optimization approach. Unlike in the original model, the intuitive global minimum variance portfolio serves as the reference portfolio. The introduction of a general rule for investors' views in combination with a simplification of the original Black-Litterman approach facilitates the implementation of the model and enables us to remove so-called dead assets from the GMV portfolio. As an additional advantageous feature our model is only based on variance-covariance estimations, and relative return estimations for our general rule. A numerical application of our modified Black-Litterman model to empirical data sets demonstrates that portfolios based on the model clearly outperform the GMV portfolio and the 1=N portfolio in terms of compound annual returns and Sharpe ratios.

Asset management companies are constantly searching for portfolio optimization models that are on the one hand clear and intuitive and on the other provide high and reliable returns. This paper presents a modifed version of the well-known Black-Litterman portfolio optimization approach. Unlike in the original model, the intuitive global minimum variance portfolio serves as the reference portfolio. The introduction of a general rule for investors' views in combination with a simplification of the original Black-Litterman approach facilitates the implementation of the model and enables us to remove so-called dead assets from the GMV portfolio. As an additional advantageous feature our model is only based on variance-covariance estimations, and relative return estimations for our general rule. A numerical application of our modified Black-Litterman model to empirical data sets demonstrates that portfolios based on the model clearly outperform the GMV portfolio and the 1=N portfolio in terms of compound annual returns and Sharpe ratios.

null

null

null

This paper develops a method to efficiently estimate hidden Markov models with continuous latent variables using maximum likelihood estimation. To evaluate the (marginal) likelihood function, I decompose the integral over the unobserved state variables into a series of lower dimensional integrals, and recursively approximate them using numerical quadrature and interpolation. I show that this procedure has very favorable numerical properties: First, the computational complexity grows linearly in time, which makes the integration over hundreds and thousands of periods well feasible. Second, I prove that the numerical error is accumulated sub-linearly over time; consequently, using highly efficient and fast converging numerical quadrature and interpolation methods for low and medium dimensions, such as Gaussian quadrature and Chebyshev polynomials, the numerical error can be well controlled even for very large numbers of periods. Lastly, I show that the numerical convergence rates of the quadrature and interpolation methods are preserved up to a factor of at least 0.5 under appropriate assumptions. I apply this method to the bus engine replacement model of Rust: first, I verify the algorithm’s ability to recover the parameters in an extensive Monte Carlo study with simulated datasets; second, I estimate the model using the original dataset.

null