The capabilities for visualization, rapid data retrieval, and manipulation in geographic information systems (GIS) have created the need new techniques of exploratory analysis that focus on “spatial” aspects data. identification local patterns spatial association is an important concern this respect. In paper, I outline a general class indicators (LISA) show how they allow decomposition global indicators, such as Moran's I, into contribution each observation. LISA statistics serve two...

Describing the various ways degree of spatial autocorrelation in a set variate values can be assessed and to which pattern formed by location objects treatable as points examined.

This article presents an overview of GeoDa™, a free software program intended to serve as user‐friendly and graphical introduction spatial analysis for non‐geographic information systems (GIS) specialists. It includes functionality ranging from simple mapping exploratory data analysis, the visualization global local autocorrelation, regression. A key feature GeoDa is interactive environment that combines maps with statistical graphics, using technology dynamically linked windows. brief...

This article outlines a taxonomy of spatial econometric model specifications that incorporate externalities in various ways. The point departure is reduced form which local or global spillovers are expressed as multipliers. From this, range familiar and less derived for the structural regression.

This paper reviews a number of conceptual issues pertaining to the implementation an explicit "spatial" perspective in applied econometrics. It provides overview motivation for including spatial effects regression models, both from theory-driven as well data-driven perspective. Considerable attention is paid inferential framework necessary carry out estimation and testing different assumptions, constraints implications embedded various specifications available literature. The review combines...

Abstract This paper reviews a number of conceptual issues pertaining to the implementation an explicit “spatial” perspective in applied econometrics. It provides overview motivation for including spatial effects regression models, both from theory‐driven as well data‐driven perspective. Considerable attention is paid inferential framework necessary carry out estimation and testing different assumptions, constraints implications embedded various specifications available literature. The review...

Several diagnostics for the assessment of model misspecification due to spatial dependence and heterogeneity are developed as an application Lagrange Multiplier principle. The starting point is a general which incorporates spatially lagged dependent variables, residual autocorrelation heteroskedasticity. Particular attention given tests in presence variables formally derived illustrated number simple empirical examples.

Based on a large number of Monte Carlo simulation experiments regular lattice, we compare the properties Moran's I and Lagrange multiplier tests for spatial dependence, that is, both error autocorrelation spatially lagged dependent variable. We consider bias power six sample sizes, ranging from twenty‐five to 225 observations, different structures weights matrix, several underlying distributions, misspecified matrices, situation where boundary effects are present. The results provide an...

Geographical and political research on urban service delivery—who benefits why—has proliferated during the past two decades. Overall, this literature is not characterized by a particular attention to importance of method in drawing conclusions about spatial equity based empirical studies. Specifically, there has been scant interest effect geographic methodology assessing relationship between access socioeconomic characteristics that are spatially defined. In paper we take analytical...

Spatial analysis is statistically and substantively important for macrolevel criminological inquiry. Using county‐level data the decennial years in 1960 to 1990 time period, we reexamine impact of conventional structural covariates on homicide rates explicitly model spatial effects. Important findings are: (1) strongly clustered space; (2) this clustering cannot be completely explained by common measures similarity neighboring counties; (3) noteworthy regional differences are observed...