Modeling spatial autocorrelation in spatial interaction data: empirical evidence from 2002 Germany journey-to-work flows

Since before the inception of work by Okabe, the intermingling of spatial autocorrelation (i.e., local distance and configuration) and distance decay (i.e., global distance) effects has been suspected in spatial interaction data. This convolution was first treated conceptually because technology and methodology did not exist at the time to easily or fully address spatial autocorrelation effects within spatial interaction model specifications. Today, however, sufficient computer power coupled with eigenfunction-based spatial filtering offers a means for accommodating spatial autocorrelation effects within a spatial interaction model for modest-sized problems. In keeping with Okabe’s more recent efforts to dissemination spatial analysis tools, this paper summarizes how to implement the methodology utilized to analyze a particular empirical flows dataset.

Hidden negative spatial autocorrelation

Mostly lip service treatments of negative spatial autocorrelation (NSA) appear in the literature, although spatial scientists confront it in practice. NSA was detected serendipitously in recalcitrant empirical analyses containing a sizeable amount of global positive spatial autocorrelation (PSA) unaccounted for by standard spatial statistical models, and labeled hidden because conventional spatial statistical tools detected only PSA while giving absolutely not hint of NSA existing. The meaning of this phenomenon is explored empirically, with findings including: a better understanding of NSA, spatial filter model construction guidelines, effective illustrations of NSA, and how hidden NSA furnishes a diagnostic for model misspecification.

Show me the code: spatial analysis and open source

This paper considers the intersection of academic spatial analysis with the open source revolution. Its basic premise is that the potential for cross-fertilization between the two is rich, yet some misperceptions about these two communities pose challenges to realizing these opportunities. The paper provides a primer on the open source movement for academicians with an eye towards correcting these misperceptions. It identifies a number of ways in which increased adoption of open source practices in spatial analysis can enhance the development of the next generation of tools and the wider practice of scientific research and education.

Hierarchical spatial interaction among the Italian regions: a nonlinear relative dynamics approach

Regional interactions and spillover effects should be considered as important factors in growth analysis of regional economies. Using modified versions of the Dendrinos–Sonis model, this paper analyses the spatial hierarchical system of Italy. The interaction among Italian regions is considered at three different levels of spatial aggregation, the NUTS-1, NUTS-2 and NUTS-3 levels. Compared to more popular spatial econometric approaches, the Dendrinos–Sonis model and its extensions provide greater flexibility in the way interaction between regions is handled but the results strongly depend on the choice of a reference region.

An exploratory analysis of hierarchical spatial interaction: the case of regional income shares in Indonesia

Regional interaction is generally understood as implying movement between regions at the same level of an hierarchy. This paper extends the notion to include an hierarchical system, thus facilitating the consideration of vertical interaction in the analysis of regional interaction. Obviously, vertical interaction is not altogether a new concept. One could find this concept in many analyses related to national-local or federal-state relationships. This paper treats hierarchy in a strict sense; spatial units are interacting one with another within, but not between, presumed super regions. A particular example drawn in this paper is the province-region relationships for the case of Indonesia. In this framework, provinces form a region, and regions form the nation. The Dendrinos–Sonis model is then used as the basis for measuring the hierarchical spatial interaction in Indonesia. The analysis will explore the degree to which complementarity and competitive interaction revealed at one level in the hierarchy persist at lower or higher levels.

Critical network infrastructure analysis: interdiction and system flow

Effective management of critical network infrastructure requires the assessment of potential interdiction scenarios. Optimization approaches have been essential for identifying and evaluating such scenarios in networked systems. Although a primary function of any network is the distribution of flow between origins and destinations, the complexity and difficulty of mathematically abstracting interdiction impacts on connectivity or flow has been a challenge for researchers. This paper presents an optimization approach for identifying interdiction bounds with respect to connectivity and/or flow associated with a system of origins and destinations. Application results for telecommunications flow are presented, illustrating the capabilities of this approach.

---

Modeling network autocorrelation within migration flows by eigenvector spatial filtering

Although the assumption of independence among interaction flows frequently is engaged in spatial interaction modeling, in many circumstances it leads to misspecified models and incorrect inferences. An informed approach is to explicitly incorporate an assumed relationship structure among the interaction flows, and to explicitly model the network autocorrelation. This paper illustrates such an approach in the context of U.S. interstate migration flows. Behavioral assumptions, similar to those of the intervening opportunities or the competing destinations concepts, exemplify how to specify network flows that are related to particular origin–destination combinations. The stepwise incorporation of eigenvectors, which are extracted from a network link matrix, captures the network autocorrelation in a Poisson regression model specification context. Spatial autocorrelation in Poisson regression is measured by the test statistic of Jacqmin-Gadda et al. (Stat Med 16(11):1283–1297, 1997). Results show that estimated regression parameters in the spatial filtering interaction model become more intuitively interpretable.

A spatiotemporal analysis of aggregate labour force behaviour by sex and age across the European Union

This study investigates the causes of variation in age-specific male and female labour force participation rates using annual data from 154 regions across ten European Union member states for the period 1983–1997. Regional participation rates appear to be strongly correlated in time, weakly correlated in space and to parallel their national counterparts. An econometric model is designed consistent with these empirical findings. To control for potential endogeneity of the explanatory variables, we use an instrumental variables estimation scheme based on a matrix exponential spatial specification of the error terms. Many empirical studies of aggregate labour force behaviour have ignored population distribution effects, relying instead on the representative-agent paradigm. In order for representative-agent models to accurately describe aggregate behaviour, all marginal reactions of individuals to changes in aggregate variables must be identical. It turns out that this condition cannot apply to individuals across different sex/age groups.

A weighted difference barrier method in landscape genetics

Identifying barriers of species and characterize their effects on spatial distribution provide essential information to research in landscape genetics. We propose a weighted difference barrier (WDB) method as an alternative to maximum difference barriers (MDB), and to initiate and integrate more spatial modeling and methods into the problem solving process. Overall, WDB provides quick and straightforward improvements to the drawbacks of MDB. WDB integrates more sample location relationships into the barrier construction and reveals potential barriers that would otherwise go undetected. WDB incorporates both within group and between group genetic information, and delineates the barriers as a more complex pattern.

A copula-based closed-form binary logit choice model for accommodating spatial correlation across observational units

This study focuses on accommodating spatial dependency in data indexed by geographic location. In particular, the emphasis is on accommodating spatial error correlation across observational units in binary discrete choice models. We propose a copula-based approach to spatial dependence modeling based on a spatial logit structure rather than a spatial probit structure. In this approach, the dependence between the logistic error terms of different observational units is directly accommodated using a multivariate logistic distribution based on the Farlie-Gumbel-Morgenstein (FGM) copula. The approach represents a simple and powerful technique that results in a closed-form analytic expression for the joint probability of choice across observational units, and is straightforward to apply using a standard and direct maximum likelihood inference procedure. There is no simulation machinery involved, leading to substantial computation gains relative to current methods to address spatial correlation. The approach is applied to teenagers’ physical activity participation levels, a subject of considerable interest in the public health, transportation, sociology, and adolescence development fields. The results indicate that failing to accommodate heteroscedasticity and spatial correlation can lead to inconsistent and inefficient parameter estimates, as well as incorrect conclusions regarding the elasticity effects of exogenous variables.

A sampling approach to estimate the log determinant used in spatial likelihood problems

Likelihood-based methods for modeling multivariate Gaussian spatial data have desirable statistical characteristics, but the practicality of these methods for massive georeferenced data sets is often questioned. A sampling algorithm is proposed that exploits a relationship involving log-pivots arising from matrix decompositions used to compute the log determinant term that appears in the model likelihood. We demonstrate that the method can be used to successfully estimate log-determinants for large numbers of observations. Specifically, we produce an log-determinant estimate for a 3,954,400 by 3,954,400 matrix in less than two minutes on a desktop computer. The proposed method involves computations that are independent, making it amenable to out-of-core computation as well as to coarse-grained parallel or distributed processing. The proposed technique yields an estimated log-determinant and associated confidence interval.

Using home buyers’ revealed preferences to define the urban–rural fringe

The location of new homes defines the urban–rural fringe and determines many facets of the urban–rural interaction set in motion by construction of new homes in previously rural areas. Home, neighborhood and school district characteristics play a crucial role in determining the spatial location of new residential construction, which in turn defines the boundary and spatial extent of the urban–rural fringe. We develop and apply a spatial hedonic variant of the Blinder (J Hum Resour 8:436–455, 1973) and Oaxaca (Int Econ Rev 9:693–709, 1973) price decomposition to newer versus older home sales in the Columbus, Ohio metropolitan area during the year 2000. The preferences of buyers of newer homes are compared to those who purchased the nearest neighboring older home located in the same census block group, during the same year. Use of the nearest older home purchased in the same location represents a methodology to control for various neighborhood, social–economic-demographic and school district characteristics that influence home prices. Since newer homes reflect current preferences for home characteristics while older homes reflect past preferences for these characteristics, we use the price differentials between newer and older home sales in the Blinder–Oaxaca decomposition to assess the relative significance of various house characteristics to home buyers.

The ecological fallacy in a time series context: evidence from Spanish regional unemployment rates

The ecological fallacy (EF) is a common problem regional scientists have to deal with when using aggregated data in their analyses. Although there is a wide number of studies considering different aspects of this problem, little attention has been paid to the potential negative effects of the EF in a time series context. Using Spanish regional unemployment data, this paper shows that EF effects are not only observed at the cross-section level, but also in a time series framework. The empirical evidence obtained shows that analytical regional configurations are the least susceptible to time effects relative to both normative and random regional configurations, while normative configurations are an improvement over random ones.

An ANN–RTS smoother scheme for accurate INS/GPS integrated attitude determination

Digital mobile mapping, the method that integrates digital imaging with direct geo-referencing, has developed rapidly over the past 15 years. The Kalman filter (KF) is considered an optimal estimation tool for real-time INS/GPS integrated kinematic positioning and orientation determination. However, the accuracy requirements of general mobile mapping applications cannot be easily achieved even when using the KF scheme. Therefore, this study proposes an intelligent scheme combining ANN and RTS backward smoother to overcome the limitations of KF and to enhance the overall accuracy of attitude determination for tactical grade and MEMS INS/GPS integrated systems.

Area coverage maximization in service facility siting

Traditionally, models for siting facilities in order to optimize coverage of area demand have made use of discrete space representations to efficiently handle both candidate facility locations and demand. These discretizations of space are often necessary given the linear functional forms of many siting models and the complexities associated with evaluating continuous space. Recently, several spatial optimization approaches have been proposed to address the more general problem of identifying facility sites that maximize regional coverage for the case where candidate sites and demand are continuously distributed across space. One assumption of existing approaches is that only demand falling within a prescribed radius of the facility can be effectively served. In many practical applications, however, service areas are not necessarily circular, as terrain, transportation, and service characteristics of the facility often result in irregular shapes. This paper develops a generalized service coverage approach, allowing a sited facility to have any continuous service area shape, not simply a circle. Given that demand and facility sites are assumed to be continuous throughout a region, geometrical properties of the demand region and the service facility coverage area are exploited to identify a facility site to optimize the correspondence between the two areas. In particular, we consider the case where demand is uniformly distributed and the service area is translated to maximize coverage. A heuristic approach is proposed for efficient model solution. Application results are presented for siting a facility given differently shaped service areas.

Nonlinear regional economic dynamics: continuous-time specification,estimation and stability analysis

This paper presents an innovative approach to the study of regional economic dynamics within a nonlinear continuous-time econometric framework—a generalized specification of the Lotka–Volterra system of equations. This specification, which accounts for interdependent behavior of three industrial sectors and spillover effects of activities in neighboring regions, is employed in an analysis of five Italian regions between 1980 and 2003. For these regions, we report estimation results, characterize the varying systems dynamics, analyze the models’ local and global stability properties, and determine via sensitivity analyses which structural features appear to exert the greatest influence on these properties.

Forecasting the impact of transport improvements on commuting and residential choice

This paper develops a probabilistic, competing-destinations, assignment model that predicts changes in the spatial pattern of the working population as a result of transport improvements. The choice of residence is explained by a new non-parametric model, which represents an alternative to the popular multinominal logit model. Travel times between zones are approximated by a normal distribution function with different mean and variance for each pair of zones, whereas previous models only use average travel times. The model’s forecast error of the spatial distribution of the Dutch working population is 7% when tested on 1998 base-year data. To incorporate endogenous changes in its causal variables, an almost ideal demand system is estimated to explain the choice of transport mode, and a new economic geography inter-industry model (RAEM) is estimated to explain the spatial distribution of employment. In the application, the model is used to forecast the impact of six mutually exclusive Dutch core-periphery railway proposals in the projection year 2020.

Income distribution dynamics and cross-region convergence in Europe

This paper presents a continuous version of the model of distribution dynamics to analyse the transition dynamics and implied long-run behaviour of the EU-27 NUTS-2 regions over the period 1995–2003. It departs from previous research in two respects: first, by introducing kernel estimation and three-dimensional stacked conditional density plots as well as highest density regions plots for the visualisation of the transition function, based on Hyndman et al. (J Comput Graph Stat 5(4):315–336, 1996), and second, by combining Getis’ spatial filtering view with kernel estimation to explicitly account for the spatial dimension of the growth process. The results of the analysis indicate a very slow catching-up of the poorest regions with the richer ones, a process of shifting away of a small group of very rich regions, and highlight the importance of geography in understanding regional income distribution dynamics.

Using fuzzy logic for modeling aquifer architecture

Modeling the geologic architecture of an aquifer and visualizing its three-dimensional structure require lithologic data recorded during well drilling. Uncertainties in layer boundaries arise due to questionable quality of drilling records, mixing during the drilling process, which results in blurred contacts, and natural heterogeneity of the geologic materials. An approach for modeling and visualizing the spatial distribution of aquifer units three-dimensionally based on fuzzy set theory is developed. An indicator is defined for evaluating the possibility of aquifer existence based on fuzzy set theory and probability principles. A specific interpolation method for aquifer 3D spatial distribution requiring only very basic borehole log data is proposed. A 3D modeling and visualization system for aquifers is also developed, which can implement basic GIS functions, like borehole identification and cross-section creation. The methodology developed is tested using real borehole lithology data available for an aquifer in British Columbia, Canada.

---

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.