Risk Measurement of Portfolios (Risikomessung von Portfolios)

By Reto Baumann / ETH Zürich , January, 2005

This thesis analyzes five different methods of risk measurement for portfolios and illustrates their practical implementation.

Objective

The objective is to identify methods that provide the best possible predictive value about the risk of a portfolio.

Conclusions

The study shows that – unlike in the case of return – it is misguided to believe that there is a single, clear-cut, exact, precise and correct measurement method for risk. Risk figures that are presented cannot be blindly trusted and must be judiciously questioned. The results are influenced by many factors, such as the choice of the measurement method, the model, the risk weighting and the historical time period reviewed. Hence, the comparison of risk information from various sources is highly questionable.

In view of this, the study attempts to provide readers with a toolbox that will enable them to carry out their own analyses efficiently and flexibly.

At the same time, however, it is worth considering which of the presented methods should be built into the risk management system of a company on a standard basis. The study provides a good overview of the applicability as well as the advantages and disadvantages of the different methods in the various situations. The approaches should by no means be implemented in an arbitrary manner. Quite the contrary. It is decisive for small and mid-sized companies to use their limited resources in a highly targeted way. In the spirit of innovation management, the most promising ideas, namely those with the best cost/benefit ratio, should be identified and implemented.