Papers by subject
Risk and Return in General: Theory and Evidence (pdf, html)
Falken Fund performance
Mutual Funds, Idiosyncratic Variance, and Asset Returns, PhD. Dissertation, Northwestern University, 1994
This dissertation documents two new facts that it argues are related. First, that accounting for various ommited variables, and measurement biases, volatility is inversely correlated with equity returns, cross-sectionally. Secondly, that mutual fund have a distinct preference towards highly volatile stocks. A simple suggestion is that volatile stocks have qualities investors like, which drives down their equilibrium returns.
"Credit Scoring for Public Companies", Credit Ratings, RiskWaters Publishing, Michael Ong Editor, 2002.
This provides an outline as to how to model default risk. Very practical.
"Tesing for Consistency in Default Rates", Journal of Fixed Income, October 2001. With Richard Cantor.
"Some Empirical Results on the Merton Model ," Risk Professional, April 2001. With Andrew Boral
Here I argue (w/ Andrey Boral) that the Merton model is not a sufficient measure of risk, and can be improved by added 'outside-the-box' information such as NI/A.. Note we used 1,450 defaults since 1980 in the US.
"RiskCalc for Private Companies 2: More Results and the Australian Model" Moody's Investors Service December 2000.
More succinct, totally explicit description of Moody's private firm default model. Examination and discussion of Australian, Canadian, and US financial ratios and their relation to default rates.
"Building statistical models for middle-market credit risk" Risk Professional, July/August 2000.
One of the key barriers to extending credit risk modelling to loan portfolios is the lack of a benchmark comparable to credit ratings in the securitised debt sector. I outline how RiskCalcÔ can help fill this role.
"Similarities and Differences between Public and Private Firm Risk " Journal of Commercial Lending, May 2000.
Can a credit analysis model estimated on public firms generate valid inferences about default probability for private firms? A discussion of the ratios of public and private firms and their relationship to default.
"RiskCalc for Private Companies: Moody's Default Model" Moody's Investors Service May 2000.
Moody's proprietary model for default risk for private firms.
"Internal Credit Risk Rating Systems Must Evolve to Stay Relevant" Journal of Lending and Credit Risk Management, May 2000.
"Validating Commercial Risk Grade Mappings: Why and How" Journal of Lending and Credit Risk Management, February 2000.
Why and how commercial risk grade buckets should be validated. This implies an empirical mapping into default and loss probabilities, which is useful for securitization, regulatory approval of internal risk grades, and general best practices.
Nassim Taleb's Black Swan, 10/07
"Shleiffer's Inefficient Markets", Book Review. 2001. Unpublished (failed webzine!).
The Hedge Fund Advantage in Equities, Financial Engineering News, 10/05
Corporate Bonds: No Reward for High Risk, RiskCenter.com article, 12/03
Lessons from the Kidder-Jett Debacle of -94, RiskCenter.com article, 10/03
Value-at-Risk and Capital Allocation, RiskCenter.com article, 10/03
Operational Risk -- Alpha Deception, RiskCenter.com article, 1/03
CEO Incompetence and Operational Risk, RiskCenter.com article, 10/03
"Risk Manager of the Future: Scientist or Poet?", Journal of Lending and Credit Risk Management, Februar 2001
"Capital Priorities: Practical Advice on Implementing RAROC" Journal of Lending and Credit Risk Management, May 1999.
The most important priority in implementing RAROC for effective balance sheet management is calculating expected losses, not portfolio modeling. A concrete example of a commercial lending model that was developed in response to this need is given in this article.
"Staying on Track" Risk Magazine, Enterprise-wide Risk Management , November 1998.
"Integrating Quantitative Risk Management Operations Across the Institution," Bank Accounting and Finance, Fall 1998.
This is an html document outlining how economic risk capital estimation can integrate firm-wide risk management within a bank. It discusses specific applications and estimation problems, highlighting the importance of nitty-gritty details in the estimation procedure.
"Value-at-Risk and Derivatives Risk," Derivatives Quarterly, Fall 1997.
This article provides empirical and theoretical support for the argument that explicitly measured credit and market risk are not the most relevant risks to a trading operation. Operating risk, a composite of all "other" risks, is the main reason trading firms fail or experience disasters. While this may seem to imply that VaR misses the point, in fact monitoring VaR can act as a means to the end of minimizing operating risks.
"Accounting for Economic and Regulatory Capital in RAROC Analysis," Bank Accounting and Finance, Fall 1997
For a bank trying to asses risk-adjusted returns on capital (RAROC) for different lines of business, understanding the differences between regulatory and economic capital needs is of prime importance. Accounting for both entails knowing which requirement is binding, and this article gives practical examples and implications of these dual requirements.
Banking Risk Website
When I was a risk manager at KeyCorp, I wrote this up on banking risk issues circa 1999
When Hedging Makes a Portfolio Worse, RiskCenter.com article, 11/03
"Model Selection and Hedging Ratios," Financial Engineering News, December 1997.
This paper was the result of a vain attempt to model interest rate caps and swaptions with the Hull-White model, as opposed to a simple Black model. It didn't work, but provides a cautionary tale when trying to implement the latest and greatest models for a bunch of traders. Bottom line: a foolish consistency is the hobgoblin of small-minded quants.
"Minimizing Basis Risk From Nonparallel Shifts in the Yield Curve Part II: Principal Components," Journal of Fixed Income June 1997
This article compares principal components-based hedges with covariance-based key rate duration hedges. It finds very similar results for the methods with one-or two principal components, but deteriorating performance for the third component, indicating significant instability of higher-order parameter estimates in principal components models.
"Minimizing Basis Risk From Nonparallel Shifts in the Yield Curve," Journal of Fixed Income, June 1996.
The aim is to construct a hedge that minimizes the variance of the hedged portfolio within a class of hedges, given the covariances among the hedging securities and the portfolio. The key-rate covariance hedge provides a significant increase in hedge efficiency over duration or yield beta approaches, and does not require highly sophisticated statistics.