Research keywords: Econometrics and Macroeconomics
George Kapetanios [82KB]
He works in the area of econometrics and macroeconomics. In the area of econometrics he is interested in (i) the analysis of nonlinear econometric models, (ii) nonlinear unit root tests, (iii)
factor and other models for large datasets (iv) model selection (v) tests of rank (vi) tests of nonlinearity, (vii) econometric forecasting and (vii) modelling under structural change. He has developed unit root tests that are powerful against nonlinear stationary processes for a variety of nonlinear alternative hypotheses. He has developed new tests for nonlinearity with very favourable size and power properties. He has proposed a new methodology for estimating factors from large datasets using state space models. In the area of macroeconomics he has investigated the persistence properties of a number of macroeconomics series and used state space and nonlinear models to forecast GDP and other macroeconomic variables for Europe and the US.
Nonlinear Models with Strongly Dependent Processes and Applications to Forward Premia and Real Exchange Rates; with R. Baillie. Journal of Econometrics; 147(1), pp. 60--71, (2008).
Panels with non-stationary multifactor error structures; with M. H. Pesaran and T. Yamagata. Journal of Econometrics, 160(2), 326--348, (2011).
Adaptive Forecasting in the Presence of Recent and Ongoing Structural Change; with L. Giraitis and S. Price. Journal of Econometrics, 177(2), 153-170, (2013).
Inference on Stochastic Time-Varying Coecient Models; with L. Giraitis and T. Yates. Journal of Econometrics, 179(1), 46-65, (2014).
A Nonlinear Panel Data Model of Cross-Sectional Dependence; with J. Mitchell and Y. Shin. Journal of Econometrics, 179(2), 134-157, (2014).