Three-Way Damage
Do you remember that cute VW video, “The Force”? The one where the little boy, dressed up as Darth Vader, can’t get a reaction from anyone or anything, even his dog, when he tries his super-powers on them? Then, Dad drives up in his VW Passat and the Vader-wannabe is, magically, able to get the lights to flash. The camera pans to parents looking out the window. Dad gives a conspiratorial grin as he clicks the remote control. On September 18, 2015, world markets were rocked by news that Volkswagen had knowingly falsified its emissions of toxic nitrogen oxides. The […]
Topology Can Streamline Modelling
How can software systematize and optimize the routine tasks in building financial risk models? “We use topology to inform feature selection, and then we examine a range of models,” said Mukund Ramachandran, Data Scientist at Ayasdi. He was the third of three panellists at the October 27, 2015, webinar on Effective Risk Models Using Machine Intelligence sponsored by the Global Association of Risk Professionals. In the course of evaluating potential models, several statistical tests are applied. “Machine intelligence considers the entire high dimensional space jointly,” he said. Machine learning is capable of applying hundreds of algorithms and different combinations, “but […]
Machine Intelligence + Business Intuition
Given the exponential growth in data complexity, how can you, the risk manager, quickly determine the most salient economic factors to include in calculating a bank’s risk exposure? Nowadays, modelling risk is all about “speed, accuracy, and defensibility,” said Patrick Rogers, Head of Marketing at the software company Ayasdi. Risk models must be developed “in a relatively short time window and must be statistically valid.” Since risk models must be defensible to business owners and industry regulators, and simple to explain, the ordinary “black box” machine learning would fall flat, Rogers said. He was the second of three panellists at […]
“Tons of Models, Tons of Variables”
With so many economic variables, and such a wide choice of parameters, do you feel overwhelmed by the task of producing the best financial model possible? Is there a systematic approach to exploring models? “Ever since the 2008 financial crisis, there’s been a focus on stress testing,” which requires robust financial models, said Roderick Powell, Director of Market and Treasury Risk at the consulting company KPMG. He was the first of three panellists at the October 27, 2015, webinar on Effective Risk Models Using Machine Intelligence sponsored by the Global Association of Risk Professionals. “Building those models is a time-consuming, […]
“Not Only The What But The How”
When it comes to financial data for stress testing, there’s a good news-bad news aspect. The good news may be that a bank did not suffer severe financial stress but the bad news is that it will be harder for the bank to model “bad events” if it does not have such data. And banks “will get written up if [the regulators] don’t believe their bad events,” said Tara Heusé Skinner, Manager at SAS Risk Research & Quantitative Solutions, and co-author of The Bank Executive’s Guide to Enterprise Risk Management. She was the first presenter of two at the May […]
Tailoring Risk Model to Investment Strategy
Due to the growing complexity of measuring financial risk, “risk has become a patchwork” of different models, said Phil Jacob, Senior Director at Axioma Risk Research. He was the sole presenter in a webinar about tailoring the right risk model to your investment strategy held on March 4, 2015, and sponsored by the Global Association of Risk Professionals (GARP). Jacob identified four inherent challenges. “There are operational issues stemming from existing rigid approaches,” leading to “difficulty in aggregating risk.” There is a lack of consistency in modeling portfolios, which can run the gamut from very simple proxies all the way […]
Stress Testing Mortgages. Part 2
The team of Scott L. Smith, Jesse Weiher, and Debra Fuller at the Federal Housing Finance Agency (FHFA) use specialized financial models to estimate potential losses. They carried out empirical tests of countercyclical shocks using four different models of mortgage credit risk. This posting continues a February 4, 2015, presentation by Scott L. Smith to an audience of financial risk managers at Global Association of Risk Professionals (GARP). Two models were devised at FHFA, and two are commercially available credit models: one, called Black Knight (formerly LPS-AA), and the other called ADCO Loan Dynamics. The estimated losses were converted to a capital […]
Stress Testing Mortgages. Part 1
“One needs to be careful and not over-reliant on any one model,” said Scott L. Smith, Associate Director for Capital Policy at the Federal Housing Finance Agency (FHFA). He was referring to the financial models used by major financial institutions to estimate potential losses. On February 4, 2015, he was presenting a GARP-sponsored webinar on countercyclical stress tests to set capital requirements. Smith explained how credit risk is measured for mortgages, and described a way to embed stress testing that uses countercyclical concepts. He and colleague Jesse Weiher, Senior Economist at FHFA, performed dynamic stress testing that was adjusted to […]
Correlation Risk
“Before we argue about correlation, we must first agree on which interpretation we are talking about,” said Gunter Meissner, President of Derivatives Software, Founder and CEO of Cassandra Capital Management, and Adjunct Professor of Mathematical Finance at NYU-Courant. He was sole presenter at a webinar on October 21, 2014, sponsored by GARP. Meissner cited three different interpretations commonly used for correlation risk. “In trading practice, it can mean similar movement in time. Or it can be narrowly defined,” he said, “to only refer to the linear Pearson definition.” Third, it can be used in the broader sense of any type […]
Clickable Calculus
When finding a definite integral, do you spend an inordinate amount of time in the step-by-step algebra? Let’s say you are integrating over a probability of default function that has been fitted to real-life data (a non-normal curve) and you want to understand step-size dependence. Or perhaps you are a beginning student of mathematical finance, reviewing the fundamentals of integration, and you just wish there was a faster way to change functions and spit out a graph. “Integration is a summative process, and the applications that show this can become a time-sink,” said Robert Lopez, Emeritus Professor of Mathematics at […]