Risks of Financial Institutions.

• Position: Bureau News

The NBER's Project on Risks of Financial Institutions met in Cambridge on November 30. Mark Carey, Federal Reserve Board, and Rene Stulz, NBER and Ohio State University, organized this program:

Torben G. Andersen, NBER and Northwestern University; Tim Boilerslev, NBER and Duke University; Francis X. Diebold, NBER and University of Pennsylvania; and Paul Labys, University of Pennsylvania, "Modeling and Forecasting Realized Volatility" (NBER Working Paper No. 8160)

Discussant: Til Schuermann, Federal Reserve Bank of New York

William Fung, London School of Business, and David Hsieh, Duke University, "The Risk in Hedge Fund Strategies: Theory and Evidence from Fixed Income Traders"

Discussant: Barry Schachter, Caxton Associates, LLC

James R. Barth, Auburn University, Gerard Caprio, Jr., World Bank, and Ross Levine, University of Minnesota, "Bank Regulation and Supervision: What Works Best?"

Discussant: Charles W Calomiris, NBER and Columbia University

Jeremy Berkowitz, University of California at Irvine, and James O'Brien, Federal Reserve Board, "How Accurate are Value-at-Risk Models at Commercial Banks?"

Discussant: Kenneth Froot, NBER and Harvard University

1. Sinan Cebenoyan, Hofstra University, and Philip E. Strahan, Boston College, "Risk Management, Capital Structure, and Lending at Banks"

Discussant: Mark Flannery, University of Florida

Andersen, Bollerslev, Diebold and Labys provide a general framework for integration of high-frequency intraday data into the measurement, modeling, and forecasting of daily and lower frequency volatility and return distributions. They formally develop the links between the conditional covariance matrix and the concept of realized volatility. Next, using continuously recorded observations for the deutschemark/dollar and yen/dollar spot exchange rates covering more than a decade, they find that forecasts from a simple long-memory Gaussian vector autoregression for the logarithmic daily realized volatilities perform admirably compared to popular daily ARCH and related models. Moreover, the vector autoregressive volatility forecast, coupled with a parametric log-normal-normal mixture distribution implied by the theoretically and empirically grounded assumption of normally distributed standardized returns, gives rise to well-calibrated density forecasts of future returns, and correspondingly accurate quantile esti mates. Their results hold promise for practical modeling and forecasting of the large covariance...