Financial risk management started in a period when academic finance was wedded to probability. Risk and its transferability was the focus and uncertainty was sidelined. After the recent financial crisis, uncertainty and its consequences have become a major concern for many prominent academics, yet practitioners are constrained by probability-based tools and regulatory mandates. Managing Uncertainty, Mitigating Risk offers a liberated perspective on uncertainty in banking and finance. The book stresses that uncertainty must be confronted by using a broader range of inputs, employing methods outside conventional probability. More often than not, systemic risks are not completely unforeseeable and a range of likely risk scenarios can be fleshed out, quantified and largely mitigated. We can accomplish this only if we widen our knowledgebase to include qualitative data and judgment. Probability and historical data alone cannot sufficiently model game-changing and catastrophic one-off situations such as Eurozone exit and breakup, US debt ceiling, and Brexit.
This book presents a robust foundation and a novel and practical method for incorporating uncertainty into existing risk frameworks. It takes the reader beyond the realms of probability in modern finance, into imprecise probability - the mathematics of uncertainty. We introduce uncertain value-at-risk (UVaR), a measure which takes the VaR engine and enhances it using credal nets, an imprecise extension of Bayesian nets. Unlike the unjustified precision of probability-based models, UVaR helps to assesses uncertainty by incorporating expert insight through priors, with more extensive datasets. By combining a solid quantitative method with an implementation framework and cases, this book allows the reader to not only understand the solution for managing uncertain one-offs, but also to see the end-product. This is a starting point for risk practitioners to go beyond regulatory-initiated tools in order to employ their own approaches towards recognizing and managing uncertainty.
1. Definitions, Applications, Methods and Tools
2. The Mathematics of Uncertainty
3. The New Framework and Approach
4. Case studies
5. Conclusions