Markowitz optimization: correcting past out-of-sample errors
نویسنده
چکیده
Portfolio optimization inputs differ widely from their subsequent out-of-sample (OOS) values. As a result, optimized portfolios have 2 to 28 times more risk OOS than their ex-ante estimates suggest. I propose a simple solution to this problem: let the data speak for itself and pick, in real time, the correction that most reduces past OOS errors. The resulting optimized portfolios consistently outperform OOS the 1/N benchmark. More significantly, the corrected covariance matrix captures very well the OOS risk of those portfolios, including the hit rates at extreme quantiles. JEL classification: G11; G12; G17.
منابع مشابه
Markowitz versus Michaud : Portfolio Optimization Strategies Reconsidered
Several attempts have been made to reduce the impact of estimation errors on the optimal portfolio composition. On the one hand, improved estimators of the necessary moments have been developed and on the other hand, heuristic methods have been generated to enhance the portfolio performance, for instance the "resampled efficiency" of Michaud (1998). We compare the out-ofsample performance of tr...
متن کاملPrediction based mean-variance model for constrained portfolio assets selection using multiobjective evolutionary algorithms
In this paper, a novel prediction based mean-variance (PBMV) model has been proposed, as an alternative to the conventional Markowitz mean-variance model, to solve the constrained portfolio optimization problem. In the Markowitz mean-variance model, the expected future return is taken as the mean of the past returns, which is incorrect. In the proposed model, first the expected future returns a...
متن کاملPortfolio Optimization Based on Cross Efficiencies By Linear Model of Conditional Value at Risk Minimization
Markowitz model is the first modern formulation of portfolio optimization problem. Relyingon historical return of stocks as basic information and using variance as a risk measure aretow drawbacks of this model. Since Markowitz model has been presented, many effortshave been done to remove theses drawbacks. On one hand several better risk measures havebeen introduced and proper models have been ...
متن کاملApplication of Clayton Copula in Portfolio Optimization and its Comparison with Markowitz Mean-Variance Analysis
With the aim of portfolio optimization and management, this article utilizes the Clayton-copula along with copula theory measures. Portfolio-Optimization is one of the activities in investment funds. Thus, it is essential to select an appropriate optimization method. In modern financial analyses, there is growing evidence indicating the distribution of proceeds of financial properties is not cu...
متن کاملSimulation of Rainfall - Runoff Events by Applying Phase Differences Diagrams and Correcting Effective Rainfall Components
The conversion of rainfall to runoff in basins includes nonlinear relations between the complex interactions of various hydrological processes. In this study, without considering of predetermined structure, relationship between input and output system was derived individually from the nature of the data recorded. Also, the phase difference occurred between rainfall and runoff signals using c...
متن کامل