What is it about?
Most portfolio optimization approaches such as the Markowitz and CVaR models minimize a measurement of risk. We develop a model that not only minimizes risk but additionally minimizes uncertainty as measured by probability entropy.
Featured Image
Why is it important?
The models developed are novel allowing a decision maker to find a set of diversified investments to make that will minimize not only risk but also probabilistic uncertainty as measured by entropy. Initial results indicate that the models perform very well when compared to the Markowitz, CVaR, and constrained MaxiMin mdoels.
Perspectives
The models developed are novel and unique and are the first to allow a decision maker to select an appropriate set of decisions to match their risk profile with respect to the probabilistic uncertainty as measured by entropy. The decision maker can select a very conservative approach using Maximin with no uncertainty, a more risky approach of using all probabilities for the Expected Value criterion, or, with our models, somewhere between those two extremes.
David F. Rogers
University of Cincinnati
Read the Original
This page is a summary of: A generalized maximin decision model for managing risk and measurable uncertainty, IISE Transactions, May 2017, Taylor & Francis,
DOI: 10.1080/24725854.2017.1335918.
You can read the full text:
Contributors
The following have contributed to this page
