Apr 30, 2013
from 05:00 PM to 06:30 PM
|Where||Barbara White Room, Newnham College|
|Contact Name||Mette Jamasb|
|Add event to calendar||
"Long Memory Affine Term Structure Models (joint with A. Golinski)
We develop a Gaussian discrete time, essentially affine term structure model which allows for long memory. This feature reconciles the strong persistence observed in nominal yields and inflation with the theoretical properties of affine models, especially for long maturities. We discuss how long memory can naturally arise with respect to the term structure of interest rates, providing a theoretical underpinning. We estimate a two factor version of the model whereby the factors have a clear economic interpretation as real short rate and expected inflation. We find that extension to long memory factors gives a substantial improvement, with respect to conventional models, in terms of model fit and forecast errors. Specifically, it seems crucial to model the expected inflation factor as a long memory process, while we do not find evidence of high persistence in the real rate dynamics. The model carries closed-form expressions for the term structure of real interest rates, real and nominal risk term premia, and inflation risk premia. All these quantities are time varying. An accurate in-sample as well as out-of-sample analysis of the model performance is presented.
Professor Zaffaroni has a summa cum laude degree in economic statistics from Roma and holds a PhD in Econometrics from the London School of Economics. He is Professor in Financial Econometrics at Imperial College Business School.
His main research interests are financial econometrics and econometric theory as well as risk management and asset allocation. His publications include The Annals of Statistics, The Journal of Econometrics, The Journal of Time Series Analysis, The Journal of Empirical Finance, The Journal of Monetary Economics and Econometric Theory. His work includes:
estimation of parametric volatility models
long memory volatility models
risk management and asset allocation
contemporaneous aggregation of linear and volatility models
estimation of parametric linear and nonlinear factor models
model averaging and its application to forecasting and model risk
Professor Zaffaroni acts as quantitative consultant in asset and risk management and as instructor of executive courses for various financial institutions.