Invited programme

[ Plenary sessions | Invited sessions in Economic Theory| Invited sessions in Econometrics ]

Invited sessions in Econometrics

August, 25th - Sunday
11.30-13.00

ECONOMETRIC ANALYSIS OF REALISED COVARIATION: HIGH FREQUENCY COVARIANCE, REGRESSION AND CORRELATION IN FINANCIAL ECONOMICS
Neil SHEPHARD, Nuffield College, Oxford, UK (with Ole E. Barndorff-Nielsen, Aarhus University, Denmark)
Discussant: Enrique SENTANA, CEMFI - Centro de Estudios Monetarios y Financieros, Madrid, Spain

This paper analyses multivariate high frequency financial data using realised covariation. We provide a new asymptotic distribution theory for standard methods such as regression, correlation analysis and covariance. It will be based on a fixed interval of time (e.g. a day or week), allowing the number of high frequency returns during this period to go to infinity. Our analysis allows us to study how high frequency correlations, regressions and covariances change through time. In particular we provide confidence intervals for each of these quantities.



August, 26th - Monday
11.30-13.00

EQUILIBRIUM SEARCH MODELS FOR MATCHED EMPLOYER/EMPLOYEE DATA Jean-Marc ROBIN, INRA, Malakoff, France
Discussant: Zvi ECKSTEIN, Tel Aviv University, Israel

In this lecture I first survey empirical features of labor markets that reveals the existence of informational frictions about the location of available jobs. I then review a particular paradigm of equilibrium models of the labor market allowing for such frictions: the so-called equilibrium search models. I particularly emphasize two classes of equilibrium search models with on-the-job search: wage posting models and wage bargaining models, and I review recent attempts at estimating these models under the assumption that the productivity of individual matches is heterogeneous in both the worker and the firm.



August, 27th - Tuesday
11.30-13.00

MODELLING OPTIMAL INSTRUMENTS FOR DYNAMIC PANEL DATA
Manuel ARELLANO, CEMFI - Centro de Estudios Monetarios y Financieros,
Madrid, Spain
Discussant: T.B.A.

Two-step instrumental variable estimators for dynamic panel data models are considered that are asymptotically efficient under some auxiliary assumptions, but remain consistent when the assumptions are violated. Asymptotic efficiency is defined in relation to the information bound for the conditional mean specification of the model. Unlike standard GMM, optimal instruments are parameterized using a fixed number of coefficients for any value of T. Thus, the properties of the resulting estimators are not fundamentally affected by the relative dimensions of T and N.



August, 28th - Wednesday
11.30-13.00

LOCAL IDENTIFICATION IN NONSEPARABLE MODELS
Andrew CHESHER, Centre for Microdata Methods and Practice, IFS and
University College London, UK
Discussant: Whitney NEWEY, Massachusetts Institute of Technology, USA

Models in which unobserved stochastic terms are nonseparable are interesting because they permit stochastic across-individual variation in the impacts of policy interventions.
This paper shows how, under rather weak conditions, local nonparametric identification of interesting features of nonseparable models can be achieved in the presence of endogenous variation in policy instruments.
Key among the identification conditions are local quantile independence of unobserved stochastic terms and local instrumental variables, and local analogs of familiar order and rank conditions.
The identification results point directly to easily computed analog estimators which are elementary functionals of estimated quantile regression functions. The results suggest that quantile regression methods are a natural tool to employ in the study of nonseparable models.