Modeling the spatial dynamics of economic models

Saguatti, Annachiara (2013) Modeling the spatial dynamics of economic models, [Dissertation thesis], Alma Mater Studiorum Università di Bologna. Dottorato di ricerca in Economia e statistica agroalimentare, 25 Ciclo. DOI 10.6092/unibo/amsdottorato/5978.
Documenti full-text disponibili:
Documento PDF (English) - Richiede un lettore di PDF come Xpdf o Adobe Acrobat Reader
Download (1MB) | Anteprima


The advances that have been characterizing spatial econometrics in recent years are mostly theoretical and have not found an extensive empirical application yet. In this work we aim at supplying a review of the main tools of spatial econometrics and to show an empirical application for one of the most recently introduced estimators. Despite the numerous alternatives that the econometric theory provides for the treatment of spatial (and spatiotemporal) data, empirical analyses are still limited by the lack of availability of the correspondent routines in statistical and econometric software. Spatiotemporal modeling represents one of the most recent developments in spatial econometric theory and the finite sample properties of the estimators that have been proposed are currently being tested in the literature. We provide a comparison between some estimators (a quasi-maximum likelihood, QML, estimator and some GMM-type estimators) for a fixed effects dynamic panel data model under certain conditions, by means of a Monte Carlo simulation analysis. We focus on different settings, which are characterized either by fully stable or quasi-unit root series. We also investigate the extent of the bias that is caused by a non-spatial estimation of a model when the data are characterized by different degrees of spatial dependence. Finally, we provide an empirical application of a QML estimator for a time-space dynamic model which includes a temporal, a spatial and a spatiotemporal lag of the dependent variable. This is done by choosing a relevant and prolific field of analysis, in which spatial econometrics has only found limited space so far, in order to explore the value-added of considering the spatial dimension of the data. In particular, we study the determinants of cropland value in Midwestern U.S.A. in the years 1971-2009, by taking the present value model (PVM) as the theoretical framework of analysis.

Tipologia del documento
Tesi di dottorato
Saguatti, Annachiara
Dottorato di ricerca
Scuola di dottorato
Scienze economiche e statistiche
Settore disciplinare
Settore concorsuale
Parole chiave
Spatial econometrics Spatial dynamic panel data Present Value Model Monte Carlo simulation
Data di discussione
4 Luglio 2013

Altri metadati

Statistica sui download

Gestione del documento: Visualizza la tesi