Professor of Finance & Econometrics & Director of the Centre for Econometric Analysis
Giovanni Urga joined Cass Business School in July 1999 as Senior Lecturer in Financial Econometrics. He was promoted to Reader in May 2001 and to Full Professor in May 2002. He was Director of the PhD Programme and Co-ordinator of the Marie Curie Training Site in "Corporate Finance, Capital Markets and Banking" from 2002-2005. Since 1992, he is Professor of Econometrics at the Economics Department of Bergamo University (Italy). His teaching includes Advanced Financial Econometrics, Advanced Financial Modelling and Forecasting, Econometrics of Financial Markets, and Stationary and Non-Stationary Panel Data Econometrics. Formely, he was Research Fellow at London Business School (1994-1999), Visiting Lecturer at New Economic School in Moscow (1996-1999), Lecturer at Queen Mary and Westfield College in London (1992-1994), and Research Officer at the Institute of Economics and Statistics in Oxford (1991-1992). Professor Urga is referee for the Journal of Applied Econometrics, International Journal of Forecasting, The Economic Journal, Journal of Economics, Economics of Innovation and New Technology, Oxford Bulletin of Economics and Statistics, Oxford Economic Papers, Economic Modelling, Economic Systems, Journal of Comparative Economics, The Econometrics Journal, International Review of Economics and Finance, the Scandanavian Journal of Economics, Journal of Economic Dynamics and Control, Journal of International Money and Finance, Journal Money Credit and Banking, Journal of Business and Economic Statistics. GUEST EDITOR (1) Special Annals Issue of the Journal of Econometrics (2005) on "Modelling Structural Breaks, Long Memory and Stock Market Volatility"; (2) Special Issue of the Journal of Business and Economic Statistics (2007) on "Common Features in London". ASSOCIATE EDITOR: Empirical Economics
BSc in Economics (Pavia University) and PhD (Oxford).
MODELLING AND TESTING FOR JUMPS IN FINANCIAL ASSETS.
We use high frequency data (BrokerTec US Treasury data on the 2-, 5-, 10- and 30- year bonds) to examine and compare the results of alternative univariate jump tests recently proposed in the literature, as a first step to evaluate the performance of these tests. The following tests are considered: Aϊt-Sahalia and Jacod (2008), Andersen, Bollerslev and Dobrev (2007), Barndorff-Nielsen and Shephard (2005), Jiang and Oomen (2006), Lee and Mykland (2007) and Mancini (2001). We are interested in identifying which tests are likely to exhibit more power, as well as in determining how the sampling frequency affects the jump identification for different tests. Moreover, we investigate how bond prices react to different types of (scheduled/ non-scheduled) information releases.
(*) Dumitru, A. and G. URGA (2012) “Identifying Jumps in Financial Assets: A Comparison between non Parametric Jump Tests”, Journal of Business and Economic Statistics 30, 242-255.
(*) Novotny, J., Petrov, D. and G. URGA (2015) “Trading Price Jump Clusters in Foreign Exchange Markets”, Journal of Financial Markets 24, 66-92.
ASYMPTOTICS AND STRUCTURAL BREAKS IN PANEL MODELS.
We develop a novel asymptotic theory for panel models with common shocks. We also propose an estimation and testing framework for parameter instability in cointegrated panel regressions with common and idiosyncratic trends. We develop tests for structural change for the slope parameters under the null hypothesis of no structural break against the alternative hypothesis of (at least) one common change point which is possibly unknown. We derive the limiting distributions of the proposed test statistics. Monte Carlo simulations examine size and power of the proposed tests.
(*) Kao, C., Trapani, L. and G. URGA (2012) “The Asymptotic for Panel Models with Common Shocks”, Econometric Reviews 31, 390-439.
(*) Kao, C., Trapani, L. and G. URGA (2016) “Testing for Instability in Covariance Structures”. Revise and Resubmit (3rd round) in Bernoulli Journal.
IDENTIFICATION ROBUST INFERENCE IN COINTEGRATING REGRESSIONS
In cointegrating regressions, available estimators and test statistics are nuisance parameter dependent. This paper addresses this problem from an identification-robust perspective with focus on set estimation of the long-run coefficient (denoted β). We propose to invert LR-type statistics that test a specified value for β against an unrestricted or a cointegration-restricted alternative. Tests in implicit form as in Phillips (1994) are also inverted. Allowing for weak identification, we propose three methods to adequately size the considered tests: a bounds-based critical value based on Dufour (1989, 1997) and Dufour and Khalaf (2002), a data-dependent "Type 2 Robust" critical value based on Andrews and Cheng (2013), and a simulation-based method based on Dufour (2006). For two empirically relevant special cases, we provide analytical solutions to the test inversion problem using the mathematics of quadrics as in Dufour and Taamouti (2005). We conduct a simulation study to assess the properties of our proposed inference methods. In addition, we also check whether and to what degree popular estimation methods, specifically the standard Maximum Likelihood of Johansen (1995), the Fully Modified OLS (Phillips and Hansen, 1990; Phillips, 1991, 1995), the Dynamic OLS of Stock and Watson (1993), and the stationarity-test based method from Wright (2000), suffer from this problem, imposing and relaxing strong exogeneity. Simulation results can be summarized as follows. The size of DOLS and FMOLS based t-tests exceeds 90% at the identification boundary. Failure of weak-exogeneity causes severe distortions for DOLS as well as for FMOLS even when β is identified. The test from Wright (2000) is also oversized at the boundary. In contrast, even when weak exogeneity fails, all our proposed LR-based corrections have good size regardless of the identification status, and good power when β is identified.
(*) Khalaf, L. and G. URGA (2014) “Identification Robust Inference in Cointegrating Regressions”, Journal of Econometrics 182, 385-396.
THE IMPACT OF MACRO NEWS ON THE TERM STRUCTURE OF INTEREST RATES.
The evaluation of the impact of the news effects is one of the key questions in financial economics and a hot topic in recent studies of macroeconomic analysis. It may not be the act of releasing information to the market which is important, nor the (gross) information embodied in the estimate itself, rather, it is the extent to which the actual announcement differs from the expected which determines the response of the market to the new information (Kim et al. 2004). The aim of this project is to increase the knowledge of the impact of macro news, coming from scheduled macro announcements, on the US interest rates term structure.
(*) Boffelli, S. and G. URGA (2014), “Evaluating Correlations in European Government Bond Spreads”, in (eds) Perna, C. and M., Sibillo), Mathematical and Statistical Methods for Actuarial Sciences and Finance, Spring.
(*) Boffelli, S. and G. URGA (2015), “Macroannouncements, Bond Auctions and Rating Actions in the European Government Bond Spreads”, Journal of International Money and Finance 53, 148-173.
(*) Boffelli, S., Skintzi, V. D. and G. URGA (2016)“High and Low Frequency Correlations in European Government Bond Spreads and Their Macroeconomic Drivers”, Journal of Financial Econometrics (Forthcoming).
BREAKS AND LONG MEMORY PROCESSES IN ECONOMICS AND FINANCE.
We propose a fractional version of two well-known credit risk pricing structural models: the Merton and Black and Cox models. We assume that the value of the firm obeys to a Geometric Fractional Brownian Motion. Prices for the equity, the bond and credit spreads are derived and a sensitivity analysis is performed. To provide a justification for these models, an empirical analysis is carried out, which employs two different datasets: Constant Maturity Yields and Moody’s Long-Term for the period December 1992–November 2003 Corporate Bond Yield Averages and Lehman Brothers Eurodollar Indices covering the period June 1996–July 2006. Long memory properties of Treasury and corporate bond yields as well as credit spreads are thus investigated.
(*) Leccadito, a., O. Rachedi, and G. URGA (2015) “Testing for True vs. Spurious Long Memory. Some Theoretical Results and a Monte Carlo Comparison”. Econometric Reviews 34, 452-479.
- 2004 - present, Centre for Econometric Analysis, Director
- 2002 - 2005, PhD Programme, Director
- 2010 - 2010, PhD programme, Director