+1 Added to my documents.
Please be aware your selection is temporary depending on your cookies policy.
Remove this selection here

Managing uncertainty with Diversification Across Macroeconomic Scenarios (DAMS): from asset segmentation to portfolio

 

The essential

Correlation is unreliable in quantifying diversification among assets

The DAMS principle

Asset segmentation according to DAMS

Moving towards DAMS second generation

A missing factor: the global risk premium

Portfolio Management: the way we look at the markets

Portfolio Management: the way we build portfolios

Portfolio diagnosis

Recent history has provided an excellent laboratory to test the robustness of investment processes. Despite claims of diversification, most balanced portfolios and pension funds were concentrated on equity risk and, consequently, key investment decisions ultimately consisted in a single binary bet: buy or sell equity. This led to pro-cyclical returns and generated a broad debate on the effectiveness of active management in generating performance in difficult market conditions. In 2011 AMUNDI Italy decided to revise the asset allocation process starting with a reinterpretation of portfolio diversification in terms of Diversification Across Macroeconomic Scenarios (DAMS). The main ambitions of DAMS are: (i) to explain complex patterns of large investment universes in terms of a limited number of factors and (ii) to catch up the market risk premium without being exposed to specific macroeconomic dynamics and asset idiosyncratic risk. In a previous study we illustrated the DAMS principle and implications in terms of asset segmentation. The aim of this paper is to move towards a new framework for multi-asset portfolio management, what we call DAMS second generation. DAMS first generation is enriched with new concepts and tools that enable us (i) to infer market expectations on relevant macroeconomic factors (growth and inflation) and global risk premium, and (ii) to properly manage portfolios via strategic and tactical asset allocation.

- - -

Introduction

In the industry, strategic asset allocation is usually achieved by means of the Markowitz model (1952). As reported in the literature, the mean-variance framework suffers from many drawbacks, in particular significant dependence on errors in input parameters (Merton, 1980) and the lack of diversification. In Pola (2014) we provided some evidence that, in a reasonable range for the volatility of diversified allocations (from 0 to 10% annualized figures), efficient portfolios exhibit only half of the available diversification. Some advances1 in strategic asset allocation have been achieved through Bayesian approaches (Black-Litterman, 1991), Robust Asset Allocation (Meucci, 2007) and entropy techniques (Pola, 2014).

[...]

FACCHINATO Simone , CIO Amundi SGR Spa
POLA Gianni , Quantitative research at Amundi

Download this article in PDF format

Send by e-mail
Managing uncertainty with Diversification Across Macroeconomic Scenarios (DAMS): from asset segmentation to portfolio
Was this article helpful?YES
Thank you for your participation.
0 user(s) have answered Yes.