Jean H. P. Paelinck

Non-standard Spatial Statistics and Spatial Econometrics

Part 1. Non-standard spatial statistics.- 1. Introduction: spatial statistics, - 2. Individual versus ecological analyses.- 3. Statistical models for spatial data: some linkages and communalities.- 4. Frequency distributions for simulated spatially autorcorrelated random variable.- 5. Understanding correlations among spatial random variables.- 6. Spatially structured random effects: a comparison of three popular specifications.- 7. Spatial filter versus conventional spatial model specifications: some comparisons.- 8. The role of spatial of autocorrelation in prioritizing sites within a geographic landscape.- 9. General spatial statistics conclusions.- 10. References: spatial statistics (Part 1) Part 2. Non-standard spatial econometrics.- 11. Introduction: spatial econometrics.- 12. Mixed linear-logarithmetic specification for Lotka-Volterra models with endogenously generated SDLS-variables.- 13. Selecting spatial regimes by threshold analysis.- 14. Finite automata.- 15 Learning from residuals.- 16. Verhulst and Poisson distributions.- 17. QUARLIREG: qualitative regression and its application to spatial data.- 18. Filtering complexity for observational errors and spatial bias.- 19. General spatial econometrics conclusions.- 20. References: spatial econometrics (Part 2).

Qualitative Multicriteria Analysis: An Application to Airport Location

A multicriteria method, which can be based on purely qualitative information as to the elements of judgment and their relative weights and set up in a probabilistic way, is first presented briefly. Multicriteria selection is then applied to the problem of airport location, the elements of the relation being taken from Dutch studies on the second national airport. Results are compared with conclusions obtained in these studies, the outcome of the synthesis being a single probability table.

Spatial Econometric Modeling of Regional Convergence in Continuous Time

In this article, the authors use a continuous-time framework to model the potential convergence dynamics in a group of regions. They propose a model based on the classical Lotka-Volterra predator-prey system of two equations—a model originally proposed by Samuelson in 1971 to perform dynamic economic analysis—and extend the model to the case of more than two regions by introducing dependence on neighboring regions. The authors state the conditions under which the system of regions moves toward a mathematically stable point of convergence and show that the model can be seen as more general than the popular β-convergence model. Finally, they also consider statistical inference and introduce a discrete approximate solution based on simultaneous dynamic least squares to estimate the parameters of the model.

On aggregation in spatial econometric modelling

Abstract. The spatial aggregation problem – also termed the modifiable areal unit problem – has attracted regular attention in spatial statistics and econometrics. In this study econometric aggregation analysis is used to investigate the formal composition of meso-areal parameters given micro-areal underlying relations with spatial dependence. Impact on stochastic terms (possible meso-areal spatial autocorrelation) is also studied. Finally consequences for meso-areal estimation are derived, the general finding having been that spatial aggregation leads to meso-region specific parameter values, with the estimation problems this implies.

Economic convergence or divergence? Modeling the interregional dynamics of EU regions, 1985–1999

Abstract.In this paper we start from a continuous time framework derived from the classical predator-prey model in order to analyze the recent dynamics of regional evolution in the EU. The model describes a system of interrelated units obeying a complex functional dynamics that at any moment may encompass divergent forces. After briefly reviewing the modeling framework presented elsewhere (Arbia and Paelinck 2002), we consider the dynamics of per-capita income in 119 NUTS-2 European regions in the years 1985–1999, and fit the model using Simultaneous Dynamic Least Squares (Paelinck 1996b). The model is shown to fit the empirical data well. The – potential divergence – leads up to a formidable coordination task for public authorities responsible for regional policies, as they are supposed – if their aim is absolute convergence in an economic sense, meaning equalizing incomes per head – to produce at any time a consistent set of regional policies for all the regions concerned, and not only for the less developed ones.

The future of large towns

A number of important facts about the growth and decline of European cities, classified according to their sizes, are presented. They are interpreted in terms of population, households, and occupancy rates. In terms of these variables, two dynamic models are presented, the properties of which are discussed in terms of some strategic parameters.

A solution method for neo-classical location problems

Abstract This article is devoted to neo-classical location problems, based on input substitution and maximization of profits. The firms optimal location is determined by means of an iterative grid search procedure, which is clarified by means of a set of numerical examples.

Specifying a joint space- and time-lag using a bivariate Poisson distribution

Separate space- or time-lags have been considered regularly in data analyses; as space–time models are more recently being studied extensively in data analytic fashion, joint estimation of both lags has to be considered explicitly. This paper addresses this issue, taking into special consideration parametric parsimony together with specification richness; use of the bivariate Poisson frequency distribution is advocated and applied to an empirical case. The relation of this approach to random effects specifications is investigated. Data for Belgian regional products constitute the empirical case study.

The Box-Cox Transformation: Computational and Interpretation Features of the Parameters

Recently attention has returned in spatial statistics and spatial econometrics to the Box-Cox and Box-Tidwell families of power transformations (e.g., van Gastel and Paelinck, 1995; Kobayashi and McAleer, 1995). In part this renewed interest anses from geo-referenced data sets almost always falling under the heading of small sample theory. In such cases design-based inference cannot rely upon the conventional central limit theorem, and model-based inference risks confusing specification and sampling error. Power transformations offer a remedy for retaining the symmetry, constant spread, and linearity assumptions typifying conventional statistical analyses. These transformations often preserve the target parameter, and produce a change in measurement scale that better characterizes the population under study. Power transformations are monotone--changing only the distance between successive numbers along a measurement scale--tend to offer modest gains in statistical efficiency and to introduce no difficulties with the consistency of inferences.

Some new estimators in spatial econometrics

Abstract In the course of spatial econometric work the necessity arises to develop adequate parameter estimators. In what follows, three cases will be developed with the respective solutions that have been adopted.