Asset Allocation

CHAPTER 23Scenario Analysis

THE CHALLENGE

Scenario analysis requires investors to define prospective economic scenarios, assign probabilities to them, translate the scenarios into expected asset class returns, and identify the most suitable portfolio, given all these inputs. The main virtue of this process is that it is intuitive. Its main drawback is that the probabilities investors assign to the alternative scenarios are derived subjectively. These probabilities, therefore, are prone to biases and perhaps to the persuasion of the less informed. This chapter addresses this drawback by describing a scientific and empirical approach to determining the probabilities of prospective economic scenarios.

COMPARISON TO MEAN-VARIANCE ANALYSIS

Before we proceed, it may be useful to place scenario analysis within the context of mean-variance analysis. Both procedures are used to construct portfolios, but they are implemented quite differently, and they appeal to different types of investors. Nevertheless, they are conceptually similar.

Mean-variance analysis requires investors to specify the expected returns, standard deviations, and correlations of asset classes, and in turn it yields efficient portfolios, along with their expected returns and standard deviations. For a given time horizon or assuming returns are expressed in continuous units, if returns are approximately elliptically distributed,1 then each portfolio's expected return and standard deviation account for all possible ...