Risk and Uncertainty in Healthcare Finance, Investment Management, and Asset Pricing

Measuring and managing risk and uncertainty has been a ongoing challenge for academics and practitioners in finance and economics. At its core, the challenge requires to understand both the randomness in the underlying process as well as the way humans respond to it when making decisions. This thesis focuses on these challenges from three different perspectives. First, from the point of view of healthcare finance, this thesis addresses key questions that occur during the drug development process and the drug deployment process. We begin by developing a systematic, quantitative, transparent, and reproducible framework that incorporates the patient’s risk and uncertainty preferences into the regulatory and decision-making process, both theoretically and empirically, to improve the design and regulation of clinical trials. Then, we consider both the development and deployment of vaccines for emerging infectious diseases. Using the COVID-19 pandemic as a case study, we develop a quantitative method to simulate and evaluate various vaccine allocation strategies when the supply of vaccines is subject to stochastic shocks. We conclude this part of the thesis by proposing and analyzing the viability of a portfolio approach aimed to improve the risk/return trade off of investment when developing mRNA vaccine candidates for 11 emerging infectious diseases. Vaccine development is not only challenging due to the high scientific risk when developing a compound, but also due to the uncertainty in the occurrence of epidemics, leading to a lack of financial incentives for pharmaceutical firms to invest in vaccine research and development. The second part of the thesis dives into the field of empirical asset pricing. If multi-factor models are routinely used in by finance academics and practitioners to understand and quantify the risk exposures of an asset, more than 150 factors have been proposed in the asset pricing literature, constituting a “factor zoo”. This thesis develops linear and nonlinear techniques to construct latent factors from a set of 150 well-known risk factors using different types of autoencoders. We then compare the performance of these latent models to classical multi-factor models on various test assets. The final part of the thesis explores an investor’s risk profile and behavioral biases in the investment management landscape and aims to understand how different market participants and different types of individuals compare along the dimensions of risk aversion and investment style. To this end, we survey a large pool of individual investors, financial advisors, and institutional investors over three years about their investment decisions under various historical and hypothetical scenarios.

Name thesis:degree_name

Doctoral

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2022

Author dc:creator

• Ben Chaouch, Zied

Advisor dc:contributor.advisor

• Lo, Andrew W.

Statement dc:rights

• In Copyright - Educational Use Permitted
• Copyright MIT

Licence dc:rights.uri

• http://rightsstatements.org/page/InC-EDU/1.0/

Handle dc:identifier.uri

https://hdl.handle.net/1721.1/147452

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/147452