“Spatial Econometrics Advanced Institute”
Rome, 26th May - 25th June 2008

AIMS OF THE SCHOOL:
To provide a solid background to Master, PhD students and young researchers interested in the analysis of spatial data with particular reference to economic applications.

VENUE:
University of Rome “La Sapienza”, Dipartimento di Studi Geoeconomici Linguistici Statistici Storici per l’Analisi Regionale, Via del Castro Laurenziano, 9.

STRUCTURE of the SCHOOL:
* 4 weeks 3 hours teaching + 2 practical classes per day
* 70 hours of teaching + 30 hours of computer lab
* final evaluation via exams (not compulsory)
* practical experiences using Matlab routines, Spacestat and Geoda.

PRE-REQUISITES
A degree (at least three years) either in economics, mathematics, statistics, quantitative geography, regional planning or similar. A mathematical background is strongly recommended. In particular it is assumed that the candidate has essential basis in calculus, probability, statistics and econometrics. Students will be provided with precise textbook references for those who want to start the program and do not possess the pre-requisites (see below). Upon demand a pre-course on basic probability, statistics and econometrics could be organized in the week 19th to 23rd of May.

DEADLINES AND TIMETABLE:
December 15th, 2007                            Deadline for expression of interest

Week I (26th-30th May)
Theoretical Spatial Economics  (13 hours of teaching)
Instructor: Prof. Jean H. Paelinck, George Mason University, Fairfax, Virginia.
Spatial Statistics (12 hours of teaching)
Instructor: Prof. G. Arbia, University “G. D’Annunzio” of Chieti.

Week II (2nd to 6th June)
Spatial econometrics I (15 hours of teaching + 10 hours lab)
Instructor: Prof. Harry  H. Kelejian, University of Maryland, College Park, Maryland.

Week III (9th to 13th June)
Spatial econometrics II (15 hours of teaching + 10 hours lab)
Instructor: Prof. Ingmar Prucha, University of Maryland, College Park, Maryland.

Week IV (16th to 20th June)
Spatial panel data (15 hours of teaching + 10 hours lab)
Instructor: Prof. Badi H. Baltagi, Syracuse University, Syracuse, New York.

Week V (23rd to 25th June)
Final examination

FEES

The fees include attendance to the courses in Rome and tutorship for the entire period, but exclude living expenses. Fees could be reduced if the number of students exceeds 20 units. Living expenses can be quantified in about € 80-100 for accommodation (in hotel) plus meals. Accommodations in apartments will also be available at a more convenient rate. The hotels are located within walking distance from the University. Acceptance of students not attending the full course is conditional upon availability of places after all full-time students have been accepted.. Members of the Spatial Econometrics Association will receive a 100 € discount on all courses.

SUPPORT AND GRANTS
A certain number of grants will be offered by the Spatial Econometrics Association to reimburse participants on a merit basis.

NUMBER OF STUDENTS ADMITTED
From a minimum of 20 to a maximum of 30 students per year.

Expression of interest (specifying name and affiliation of the student) should be sent as soon as possible and in any case December the 15th to:  secretary@spatialeconometricsassociation.org

1. Prerequisites

Probability - Probability theory. Random variables, stochastic independence, conditional expectations and martingales. Stochastic processes. Markov processes. Stationary processes.
Reference: G.R. Grimmett and D.R. Strizaker (2001) Probability and Random Processes, 3nd ed.

Statistics -Statistical inference. Likelihood, the general principles of inference (sufficiency, conditionality, invariance). Theory of point estimation (Fisher information and efficiency of estimators, properties of Maximum Likelihood Estimators, and Bayesian point estimators), interval estimation and hypothesis testing. The three tests based on likelihood. Pseudo Maximum Likelihood Estimation.
References: Casella and R. L. Berger (2002) Statistical Inference, Second Edition, Duxbury Press, (CB); Davidson A. C. Statistical models, Cambridge university press; Pace L.  and  Salvan A. (1997) Principles of statistical inference from a neo-fisherian perspective, World Scientific; Young G.A., Smith R.L., (2005) Essentials of statistical inference, Cambridge, Cambridge University Press.

Econometrics  - Linear Regression Model, the OLS estimator and the violation of the Gauss-Markov hypotheses. The generalized linear model, Instrumental variable estimation, non linear least squares, 2SLS.
References: Davidson, R. and J. MacKinnon (1993): Estimation and Inference in Econometrics, Oxford University Press, Oxford. Greene, W. (2003): Econometric Analysis , Third Edition, Prentice Hall, Englewood Cliffs. Gouriéroux, C., and A. Monfort (1995): Statistics and Econometric Models, Volumes I and II, Cambridge University Press, Cambridge.

2. Courses offered

Theoretical Spatial Economics (J. H. P. Paelink, 13 hours)  - As not all of the potential students would have attended a class in theoretical spatial economics, this course would allow the group to get familiarized with essential theoretical notions. These would include: the structures of TSE, various distances, equilibria (comprising Tinbergen-Bos Systems), location theory, spatial pricing, spatial input-output analysis, shopping models and spatial dynamics.
References: Paelinck, J.H.P. and Nijkamp, P., (1975) Operational Theory and Method in Regional Economics, Saxon House, Farnborough. Paelinck, J.H.P., with the assistance of J.‑P. Ancot and J.H. Kuiper, (1977) Formal Spatial Economic Analysis, Gower Press, Aldershot. Paelinck, J.H.P., avec le concours de J.-P. Ancot, J.H. Kuiper et M. ten Raa, (1985) Eléments d'Analyse Economique Spatiale, Editions Régionales Européennes, Genève, et Editions Anthropos, Paris. Other, more recent material, will be made available.

Spatial Statistics (G. Arbia, 13 hours) - Point processes theory (complete spatial randomness, distance methods, k-functions), multivariate point processes, marked point processes, space-time point patterns. Random fields theory, conditional and simultaneous Gaussian fields, Markov random fields, non Markov random fields, dynamic fields, separable and non-separable space-time models. Stationary processes on a continuous space: variogram and co-variogram, the spectral representation, spatial prediction and krieging.
References: Banerjee, S., Carlin, B. P., and Gelfand, A. E. (2004). Hierarchical Modeling and Analysis for Spatial Data. Chapman & Hall/CRC, Boca Raton, FL. Cressie, N (1993) Statistics for spatial data, Wiley. Diggle, P.J. (2003). Statistical Analysis of Spatial Point Patterns (second edition). London: Edward Arnold. Diggle, P.J. and Ribeiro, P.J. Jnr (2007). Model-based Geostatistics. New York: Springer. Guyon X. (1995) Random fields on a network: modeling, statistics, and applications, Springer Verlag. Haining R P (2001) Spatial Data Analysis: Theory and Practice, Cambridge University Press.

Spatial Econometrics 1 (H. Kelejian, 15 hours) - Elements of large sample theory, single equation Cliff-Ord type models and variations, illustrations, specification, weighting matrix and parameter space issues, estimation including MLE, GMM, GLS, GS2SLS, large sample results and corresponding inferences, emanating and self feedback effects implied by the models, various estimation problems including border issues, uniform weights, and parameterized weighting matrices, a spatial J-Test of specifications.
References: Anselin, L. (1988), Spatial Econometrics: Methods and Models.  Boston: Kluwer Academic Publishers; Arbia, G. (2006), Spatial Econometrics: Statistical Foundations and Applications to Regional Growth Convergence, New York: Springer; Cliff, A. and Ord, J. (1981), Spatial Processes, Models and Applications.  London: Pion; Cressie, N.A.C. (1993), Statistics of Spatial Data. New York: Wiley; Green, W. (2003), Econometric Analysis, Englewood Cliffs: Prentice Hall; articles.

Spatial Econometrics 2 (I. Prucha, 15 hours) - Further discussion of single equation Cliff-Ord type models, efficient instruments and best GS2SLS, prediction, estimation in case of heteroskedastic innovations by MLE,  GMM, GS2SLS, large and small sample results. Simultaneous equation Cliff-Ord type models, estimation theory for limited and full information estimators. Spatial HAC variance covariance matrix estimation. Testing for spatial dependence, classical Moran I test and extensions. Recent developments towards estimation theory for nonlinear models, if time permits.
References: Anselin, L. (1988), Spatial Econometrics: Methods and Models.  Boston: Kluwer Academic Publishers; Arbia, G. (2006), Spatial Econometrics: Statistical Foundations and Applications to Regional Growth Convergence, New York: Springer; Cliff, A. and Ord, J. (1981), Spatial Processes, Models and Applications.  London: Pion; Cressie, N.A.C. (1993), Statistics of Spatial Data. New York: Wiley; Green, W. (2003), Econometric Analysis, Englewood Cliffs: Prentice Hall; articles.

Spatial panel data (B. Baltagi, 15 hours) - Panel data models: fixed effects and random effects. Temporal Heterogeneity. Spatial Seemingly Unrelated Regressions. Spatio-Temporal Models. Error Components with Space-Time Dependence. Specification of spatial panel models. Estimation of Spatial Panel Models: Maximum Likelihood Estimation, Instrumental Variables and GMM. Testing for spatial dependence in spatial panels.
References: Anselin, L, Le Gallo, J., and Jayet, J. (2007) Spatial Panel Econometrics, In L. Matyas and P. Sevestre (Eds.), The Econometrics of Panel Data, Fundamentals and Recent Developments in Theory and Practice (3rd Edition). Dordrecht, Kluwer. Baltagi, B. H. (2001). Econometric Analysis of Panel Data (Second Edition). John Wiley & Sons, Chichester, United Kingdom. Baltagi, B. H., Song, Seuck H., and Koh, W. (2003b). Testing panel data regression models with spatial error correlation. Journal of Econometrics, 117:123–150.

THE SPATIAL ECONOMETRICS ASSOCIATION
The Spatial Econometrics Association was founded on May 26th 2006 by the current members of the Board of Directors, with the aim of promoting "the development of theoretical tools and sound applications of the discipline of spatial econometrics, including spatial statistics and spatial data analysis". According to the bye-law of the Association "Spatial econometrics should be viewed in a wide sense involving developments of models and statistical tools for the analysis of externalities, spillovers, interactions etc. in various areas including economics, geography and regional science, etc. An aim of the Association is to disseminating and encourage such a knowledge and good practice in academic and research institutions and in the society at large". The SEA currently counts about 100 members coming from different continents of the world. After the founding conference, held in Rome in May 2006, the first official Conference of the Association was held in Cambridge in July 2007. In 2008 the Conference will be held New York (17-19 November). The 2009 Annual Conference is scheduled to be held in Barcellona in spring. The previous conferences gave rise to two special issues respectively appearing on the journals "Empiric Economics" (forthcoming, March 2008) and "Papers in Regional Science" (forthcoming, October 2008). The 2008 conferece will also produce a special issue of the journal "Regional Studies and Urban Economics".

Board of Directors:
Luc Anselin – Arizona State University, Phoenix, Arizona, US
Giuseppe Arbia – “D’Annunzio” University of  Pescara, IT
Badi H. Baltagi - Syracuse University, Syracuse, New York, US
Harry  H. Kelejian - University of Maryland, College Park, Maryland, US
Jean H. Paelinck – George Mason University, Fairfax, Virginia, US
Ingmar Prucha - University of Maryland, College Park, Maryland, US
Peter Robinson – London School of Economics, London, UK

More information about the SEA can be found on the website www.spatialeconometricsassociation.org