The Departments of Statistics and Economics jointly organise these seminars throughout the year. All are welcome to attend.
Please contact Dr Marcia Schafgans and Dr Matteo Barigozzi for further information or visit the Department of Economics Website
Please address general enquiries to Sarah McManus.
Friday 24 October 2014, 12pm - 1pm, Room LG.03, 32 Lincoln's Inn Fields, Department of Economics
Maps and directions
David Preinerstorfer
Universität Wien
Title: On size and power of heteroskedasticity and autocorrelation robust tests
Friday 7 November 2014, 12pm - 1pm, Room LG.03, 32 Lincoln's Inn Fields, Department of Economics
Maps and directions
Marcia Schafgans
LSE
Title: Inference and homogeneity in large dynamic panels with strong cross sectional dependence
Friday 21 November 2014, 12pm - 1pm, Room LG.03, 32 Lincoln's Inn Fields, Department of Economics
Maps and directions
Vassilis Hajivassiliou
LSE
Title: Establishing the coherency of dynamic LDV models
Friday 28 November 2014, 12pm - 1pm, Room LG.03, 32 Lincoln's Inn Fields, Department of Economics
Maps and directions
Abhimanyu Gupta
University of Essex
Title: to be confirmed
Friday 5 December 2014, 12pm - 1pm, Room LG.03, 32 Lincoln's Inn Fields, Department of Economics
Maps and directions
Senay Sakullu
University of Bristol
Title: Analysis of two-sided markets: an application to local American newspapers
Friday 30 January 2015: 12pm - 1pm, Room COL 6.15, Columbia House (sixth floor), Department of Statistics
Maps and directions
Liudas Giraitis
Queen Mary University of London
Title: Inference on stochastic time-varying coefficient models (joint with G Kapetanios and T Yates)
Abstract: Recently, there has been considerable work on stochastic time-varying coefficient models as vehicles for modelling structural change in the macroeconomy with a focus on the estimation of the unobserved paths of random coefficient processes. The dominant estimation methods, in this context, are based on various filters, such as the Kalman filter, that are applicable when the models are cast in state space representations. This paper introduces a new class of autoregressive bounded processes that decompose a time series into a persistent random attractor, a time varying autoregressive component, and martingale difference errors. The paper examines, rigorously, alternative kernel based, nonparametric estimation approaches for such models and derives their basic properties. These estimators have long been studied in the context of deterministic structural change, but their use in the presence of stochastic time variation is novel. The proposed inference methods have desirable properties such as consistency and asymptotic normality and allow a tractable studentization. In extensive Monte Carlo and empirical studies, we find that the methods exhibit very good small sample properties and can shed light on important empirical issues such as the evolution of infation persistence and the purchasing power parity (PPP) hypothesis.
Friday 6 February 2015: 12pm - 1pm, Room COL 6.15, Columbia House (sixth floor), Department of Statistics
Maps and directions
Qing Pei
University of Hong Kong
Title: Nomadic destiny and mandate of heaven: a new perspective on the nomadic migration from environmental humanities
Abstract: The push force of the nomadic migration has been closely related with climate change, but to date there are insufficient evidences to prove it. Following the paradigm of Environmental Humanities, the study investigated the relationship between a 2000-year history of the nomadic migration and climate change in historical China. By using different statistical methods and a large amount of updated data, the study solved several unanswered questions from past research about the relationship between climate change and the nomadic migration, especially over the long term and on a large spatial scale. In addition, the nomadic migration is a key factor in influencing the alternating occupancy patterns of the country’s pastoral and agrarian polities. Therefore, the long-term cyclical patterns of China’s geopolitical shifts have been further explained based on the nomadic migration under the impact of climate change.
Friday 13 February 2015: 12pm - 1pm, Room COL 6.15, Columbia House (sixth floor), Department of Statistics
Maps and directions
Karim Abadir
Imperial College London
Title: Link of moments before and after transformations, with an application to resampling from fat-tailed distributions
Abstract: Let x be a transformation of y. We derive an expansion formulating the expectations of x in terms of the expectations of y. Apart from the intrinsic interest in such a fundamental relation, our results can be applied to calculating E(x) by the low-order moments of a transformation which can be chosen to give a good approximation for E(x). To do so, we generalize the approach of bounding the terms in expansions of characteristic functions, and use our result to derive an explicit and accurate bound for the remainder when a finite number of terms are taken. We illustrate one of the implications of our method by providing accurate naïve bootstrap confidence intervals for the mean of a fat-tailed distribution with an infinite variance, in which case currently-available bootstrap methods are asymptotically invalid and unreliable in finite sample.
Keywords: Expansion of functions; Remainder's bound; Complex analysis, Moments; Bootstrap confidence interval; Infinite variance; Stable law.
Joint work with Adriana Cornea-Madeira
Friday 27 February 2015: 12pm - 1pm, Room COL 6.15, Columbia House (sixth floor), Department of Statistics
Maps and directions
Marcelo Fernandez
Queen Mary University of London
Title: Smoothing quantile regressions by Marcelo Fernandes, Emmanuel Guerre and Eduardo Horta
Abstract: The paper proposes a new smoothed version of the quantile regression estimator. A second-order approximation shows that the mean squared error of our estimator is smaller than that of the standard quantile regression estimator for optimal bandwidths. We then propose a data-driven choice of the bandwidth through cross-validation. A simulation experiment reveals a quite substantial improvement, with the mean square error reducing by as much as 40% relative to the standard quantile regression estimator.
Friday 13 March 2015: 12pm - 1pm, Room COL 6.15, Columbia House (sixth floor), Department of Statistics
Maps and directions
Davide La Vecchia
University of St Gallen and Monash University
Title: Semiparametrically Efficient R-Estimation for Dynamic Location-Scale Models
Authors: Marc Hallin and Davide La Vecchia
Abstract: The seminar talk introduces new rank-based procedures designed to conduct semiparametric inference on time series models. In the considered setting, the conditional location and scale of the process depend on an Euclidean parameter, while the innovation density is an infinite dimensional nuisance parameter. Easy-to-implement rank-based estimators (R-estimators) are derived and their properties are discussed, with emphasis on semiparametric efficiency and root-n consistency, even in the presence of misspecification. The developed methodology has a wide range of applications, including linear and nonlinear models, in either discrete- or continuous-time, with either homo- or heteroskedasticity. Numerical examples about the modeling of the two scale realized volatility illustrate the performances of the proposed R-estimators. Finally, some extensions related to (i) indirect inference, (ii) constrained inference on conditional duration models for market microstructure analysis, and (iii) multivariate time series are briefly discussed.