The Detection of Clusters Using a Spatial Version of the Chi-Square Goodness-of-Fit Statistic

A test statistic for the detection of spatial clusters is developed by generalizing the common chi-square goodness-of-fit test. The paper includes a discussion of the relationship between the statistic and other associated statistics, and provides an analysis of both its null distribution and power. The paper concludes with the development of a local version of the statistic and an application to leukemia clustering in central New York.     

USING THE GETIS STATISTIC TO EXPLORE AGGREGATION EFFECTS IN METROPOLITAN TORONTO CENSUS DATA

In an earlier paper by the same authors (Amrhein and Reynolds 1996), the variation of the G statistic (Getis 1991; Getis and Ord 1992) was demonstrated to correlate very highly with calculated aggregation effects. In the conclusions to this earlier paper, we called for experiments with different data sets, different aggregation rules, and different definitions of the connectivity matrix. In this paper, we explore the behaviour of census data from enumeration areas of the Toronto Census Metropolitan Area. Experiments are conducted that describe the aggregation effects that creep into the data at different scales, and the effect of the aggregation algorithm on the results. Finally, we test the ability of a modified G statistic, our Getis statistic, to predict aggregation effects     

Testing for Local Spatial Autocorrelation in the Presence of Global Autocorrelation

A fundamental concern of spatial analysts is to find patterns in spatial data that lead to the identification of spatial autocorrelation or association. Further, they seek to identify peculiarities in the data set that signify that something out of the ordinary has occurred in one or more regions. In this paper we provide a statistic that tests for local spatial autocorrelation in the presence of the global autocorrelation that is characteristic of heterogeneous spatial data. After identifying the structure of global autocorrelation, we introduce a new measure that may be used to test for local structure. This new statistic Oi is asymptotically normally distributed and allows for straightforward tests of hypotheses. We provide several numerical examples that illustrate the performance of this statistic and compare it with another measure that does not account for global structure.     

A Local Indicator of Multivariate Spatial Association: Extending Geary's c

This paper extends the application of the Local Geary c statistic to a multivariate context. The statistic is conceptualized as a weighted distance in multivariate attribute space between an observation and its geographical neighbors. Inference is based on a conditional permutation approach. The interpretation of significant univariate Local Geary statistics is clarified and the differences with a multivariate case outlined. An empirical illustration uses Guerry's classic data on moral statistics in 1830s France.     

OLS ESTIMATION AND THE t TEST REVISITED IN RANK‐SIZE RULE REGRESSION*

ABSTRACT The rank‐size rule and Zipf's law for city sizes have been traditionally examined by means of OLS estimation and the t test. This paper studies the accurate and approximate properties of the OLS estimator and obtains the distribution of the t statistic under the assumption of Zipf's law (i.e., Pareto distribution). Indeed, we show that the t statistic explodes asymptotically even under the null, indicating that a mechanical application of the t test yields a serious type I error. To overcome this problem, critical regions of the t test are constructed to test the Zipf's law. Using these corrected critical regions, we can conclude that our results are in favor of the Zipf's law for many more countries than in the previous researches such as Rosen and Resnick (1980) or Soo (2005) . By using the same database as that used in Soo (2005) , we demonstrate that the Zipf law is rejected for only one of 24 countries under our test whereas it is rejected for 23 of 24 countries under the usual t test. We also propose a more efficient estimation procedure and provide empirical applications of the theory for some countries.     

An Information-theoretic Analysis: An Application to Soviet-COMECON Trade Flows

This paper develops a classificatory methodology designed to assign geographic entities to groupings based upon the multivariate flow profiles displayed by the entities and a dispersion statistic designed to detect dynamic shifts in these flow profiles over time. Information statistic, provide the mathematical basis of the analysis and the changing trade profiles of the member nations of COMECON provide an illustrative case study.     

Spatial autocorrelation tests and the Classic Maya collapse: Methods and inferences

Spatial autocorrelation, resulting in pattern or structure in geographically distributed data, is discussed in theoretical and practical terms. Tests for spatial autocorrelation are presented, along with an explication of the relationship between autocorrelation models, the product-moment correlation coefficient and the spatial autocorrelation test statistic. Two archaeological examples illustrate the application of the auto-correlation test statistic. The first uses a hypothetical data set, which shows the type of map patterns that appear with various levels of spatial autocorrelation, and the second examines the terminal distribution of long-count-dated monuments at lowland Classic Maya sites. The results of the second example fail to support arguments for simple patterning in the cessation of the erection of these monuments and, by inference, in the Maya collapse itself. Finally, it is argued that while the identification of spatial autocorrelation is often the goal of spatial analyses, the presence of autocorrelation violates the assumptions of certain statistics used in such analyses.     

ONCE AGAIN: TESTING FOR REGIONAL HOMOGENEITY*

ABSTRACT. Summarizing the foregoing discussions in this journal on testing for regional homogeneity the present note shows that in the model of Zellner's seemingly unrelated regressions one test statistic may be used not only to test for overall homogeneity but also to examine for individual coefficient homogeneity. This aim is achieved by varying the linear restrictions in the test statistic according to different problems. To illustrate these tests regional consumption functions for the 11 Bundesläder (States) of the Federal Republic of Germany are used.     

A Classification of Spatial Distributions Based upon Several Cell Sizes

Several procedures, based upon cell count analysis, have been proposed for classifying spatial distributions, or maps, associated with some region R. Such procedures are rather imprecise and are known to depend upon the sixes and shapes of the cells in the particular partition of R under consideration. In this paper, the problem is considered from the point of view of hypothesis testing. A test of randomness based upon an arbitrary number of partitions of R is giuen. If the hypothesis of randomness is rejected, additional tests may be performed to classify the map into one of two categories, clustered or regular. These tests provide a number of advantages over existing procedures. Based upon multiple partitions of R, they decrease the dependence upon any particular partition, and the colresponding classification is precise since the null hypothesis distribution of the test statistic is (asymptotically) known. Finally, they allow a great deal of flexibility in testing for certain alternatives to randomness, and are applicable to one-, two-, and three- dimensional maps.     

Beyond Moran's I: Testing for Spatial Dependence Based on the Spatial Autoregressive Model

The statistic known as Moran's I is widely used to test for the presence of spatial dependence in observations taken on a lattice. Under the null hypothesis that the data are independent and identically distributed normal random variates, the distribution of Moran's I is known, and hypothesis tests based on this statistic have been shown in the literature to have various optimality properties. Given its simplicity, Moran's I is also frequently used outside of the formal hypothesis-testing setting in exploratory analyses of spatially referenced data; however, its limitations are not very well understood. To illustrate these limitations, we show that, for data generated according to the spatial autoregressive (SAR) model, Moran's I is only a good estimator of the SAR model's spatial-dependence parameter when the parameter is close to 0. In this research, we develop an alternative closed-form measure of spatial autocorrelation, which we call APLE , because it is an approximate profile-likelihood estimator (APLE) of the SAR model's spatial-dependence parameter. We show that APLE can be used as a test statistic for, and an estimator of, the strength of spatial autocorrelation. We include both theoretical and simulation-based motivations (including comparison with the maximum-likelihood estimator), for using APLE as an estimator. In conjunction, we propose the APLE scatterplot, an exploratory graphical tool that is analogous to the Moran scatterplot, and we demonstrate that the APLE scatterplot is a better visual tool for assessing the strength of spatial autocorrelation in the data than the Moran scatterplot. In addition, Monte Carlo tests based on both APLE and Moran's I are introduced and compared. Finally, we include an analysis of the well-known Mercer and Hall wheat-yield data to illustrate the difference between APLE and Moran's I when they are used in exploratory spatial data analysis.     

Power of the Rank Adjacency Statistic to Detect Spatial Clustering in a Small Number of Regions

The rank adjacency statistic D is a statistical method for assessing spatial autocorrelation or clustering of geographical data. It was originally proposed for summarizing the geographical patterns of cancer data in Scotland (IARC 1985). In this paper, we investigate the power of the rank adjacency statistic to detect spatial clustering when a small number of regions is involved. The investigations were carried out using Monte Carlo simulations, which involved generating patterned/clustered values and computing the power with which the D statistic would detect it. To investigate the effects of region shapes, structure of the regions, and definition of weights, simulations were carried out using two different region shapes, binary and nonhinary weights, and three different lattice structures. The results indicate that in the typical example of considering Canadian total mortality at the electoral district level, the D statistic had adequate power to detect general spatial autocorrelation in twenty‐five or more regions. There was an inverse relationship between power and the level of connectedness of the regions, which depends on the weighting function, shape, and arrangement of the regions. The power of the D statistic was also found to compare favorably with that of Moran's I statistic.     

Exploring Goodness-of-Fit and Spatial Correlation Using Components of Tango's Index of Spatial Clustering

The ability to detect anomalies such as spatial clustering in data sets plays an important role in spatial data analysis, leading to interest in test statistics identifying patterns exhibiting significant levels of clustering. Toward this end, Tango (1995) proposed a statistic (and its associated distribution under a null hypothesis of no clustering) assessing overall patterns of spatial clustering in a set of observed regional counts. Rogerson (1999) observed that Tango's index may be decomposed into the summation of two distinct statistics, the first mirroring standard tests of goodness-of-fit (GOF), and the second an index of spatial association (SA) similar to Moran's I . In this article, we investigate the effectiveness of Rogerson's expression of Tango's statistic in separating GOF from SA in data sets containing clusters. We simulate data under the null hypothesis of no clustering as well as two alternative hypotheses. The first alternative hypothesis induces a poor fit from the null hypothesis while maintaining independent observations and the second alternative hypothesis induces spatial dependence while maintaining fit. Using Rogerson's decomposition and leukemia incidence data from upstate New York, we show graphically that one is unable to statistically distinguish poor fit from autocorrelation.     

Comparative Spatial Filtering in Regression Analysis

One approach to dealing with spatial autocorrelation in regression analysis involves the filtering of variables in order to separate spatial effects from the variables’ total effects. In this paper we compare two filtering approaches, both of which allow spatial statistical analysts to use conventional linear regression models. Getis’ filtering approach is based on the autocorrelation observed with the use of the G_i local statistic. Griffith's approach uses an eigenfunction decomposition based on the geographic connectivity matrix used to compute a Moran's I statistic. Economic data are used to compare the workings of the two approaches. A final comparison with an autoregressive model strengthens the conclusion that both techniques are effective filtering devices, and that they yield similar regression models. We do note, however, that each technique should be used in its appropriate context.     

Spatial Entropy

A major problem in information theory concerns the derivation of a continuous measure of entropy from the discrete measure. Many analysts have shown that Shannon's treatment of this problem is incomplete, but few have gone on to rework his analysis. In this paper, it is suggested that a new measure of discrete entropy which incorporates interval size explicitly is required; such a measure is fundamental to geography and this statistic has been called spatial entropy. The use of the measure is first illustrated by application to one-and two-dimensional aggregation problems, and then the implications of this statistic for Wilson's entropy-maximizing method are traced. Theil's aggregation statistic is reinterpreted in spatial terms, and finally, some heuristics are suggested for the design of real and idealized spatial systems in which entropy is at a maximum.     

Diffusion patterns of violence in civil wars

Much of the current conflict literature attempts to explain the occurrence of violence as the result of determinants exogenous to the conflict process. This paper takes a different approach and analyzes how violence in civil wars spreads in space and time, drawing on earlier work on micro-diffusion of violence in criminology as well as high resolution conflict data. Two general scenarios are distinguished in our analysis: the relocation and the escalation of conflict. Relocation diffusion corresponds to a shift in the location of violence, whereas escalation diffusion refers to the spatial expansion of the conflict site. We argue that unconventional warfare in civil wars without demarcated front lines should primarily lead to the second type of pattern. We describe an extension to a joint count statistic to measure both diffusion types in conflict event data. Monte Carlo simulation allows for the establishment of a baseline for the frequency of contiguous conflict events under the assumption of independence, and thus provides a significance test for the observed patterns. Our results suggest that violence in civil wars exhibits patterns of diffusion, and in particular, that these patterns are primarily of the escalation type, driven by the dynamic expansion of the scope of the conflict.     

An Approximation for the Rank Adjacency Statistic for Spatial Clustering with Sparse Data

The rank adjacency statistic D provides a simple method to assess regional clustering. It is defined as the weighted average absolute difference in ranks of the data, taken over all possible pairs of adjacent regions. In this paper the usual normal approximation to the D statistic is found to give inaccurate results if the data are sparse and some regions have tied ranks. Adjusted formulae for the moments of D that allow for the existence of ties are derived. An example of analyses of sparse mortality data (with many regions having no deaths, and hence tied ranks) showed satisfactory agreement between the adjusted formulae and the empirical distribution of the D statistic. We conclude that the D statistic, when used with adjusted moments, provides a valid approximate method to evaluate spatial clustering, even in sparse data situations.     

Local Indicators of Spatial Association—LISA

The capabilities for visualization, rapid data retrieval, and manipulation in geographic information systems (GIS) have created the need for new techniques of exploratory data analysis that focus on the “spatial” aspects of the data. The identification of local patterns of spatial association is an important concern in this respect. In this paper, I outline a new general class of local indicators of spatial association (LISA) and show how they allow for the decomposition of global indicators, such as Moran's I, into the contribution of each observation. The LISA statistics serve two purposes. On one hand, they may be interpreted as indicators of local pockets of nonstationarity, or hot spots, similar to the G_i and G*_i statistics of Getis and Ord (1992). On the other hand, they may be used to assess the influence of individual locations on the magnitude of the global statistic and to identify “outliers,” as in Anselin's Moran scatterplot (1993a). An initial evaluation of the properties of a LISA statistic is carried out for the local Moran, which is applied in a study of the spatial pattern of conflict for African countries and in a number of Monte Carlo simulations.     

The Analysis of Spatial Association by Use of Distance Statistics

Introduced in this paper is a family of statistics, G, that can be used as a measure of spatial association in a number of circumstances. The basic statistic is derived, its properties are identified, and its advantages explained. Several of the G statistics make it possible to evaluate the spatial association of a variable within a specified distance of a single point. A comparison is made between a general G statistic and Moran's I for similar hypothetical and empirical conditions. The empirical work includes studies of sudden infant death syndrome by county in North Carolina and dwelling unit prices in metropolitan San Diego by zip-code districts. Results indicate that G statistics should be used in conjunction with I in order to identify characteristics of patterns not revealed by the I statistic alone and, specifically, the G_i and G_i* statistics enable us to detect local “pockets” of dependence that may not show up when using global statistics.     

Optimal Weights for Focused Tests of Clustering Using the Local Moran Statistic

Local spatial statistics measure and test for spatial association for a variable or variables of interest in a geographic neighborhood surrounding a predefined location. Most applications adopt a single scale of analysis but give little attention to the scale of the process generating the data. Alternatively, when the researcher is uncertain about the process scale, local statistics may examine a number of scales. In these cases, it is important to include a correction for multiple testing when evaluating the statistical significance of each local statistic, something that is rarely done. Consequently, local statistics are more likely to identify significant relationships, even when no meaningful spatial association exists. In this article, we develop a methodology for the local Moran statistic that provides both an empirical estimate of the spatial scale of association and an assessment of the significance of the statistic for that scale. The key idea is to test a number of possible choices for the statistic's weight matrix and then account for the multiple testing associated with the number of weight matrices examined. Unlike previous research, our statistic avoids the use of simulation to determine statistical significance in the presence of multiple testing. To test the validity of our approach, we constructed a numerical example to assess the statistic's performance and conducted an empirical study using leukemia data from central New York state. The developed statistic addresses the need for the empirical determination of weights and spatial scale. The test therefore addresses the common weakness of many applications, where weights are defined exogenously, with little or no thought given to either the definition or its implications. Los indicadores locales (local spatial statistics) evalúan la asociación espacial de una o varias variables de interés dada un área predefinida y sus áreas vecinas. La mayoría de dichas medidas utilizan una escala única de análisis y prestan poca atención a la escala del proceso de generación de los datos. En los casos en los que el investigador no está seguro de la escala del proceso, las los indicadores locales pueden ser evaluados a varias escalas. En dichos casos, cuando se hace la evaluación de la significancia estadística de cada indicador local, es importante incorporar una corrección para pruebas múltiples (multiple tests), un ajuste que raramente se realiza en la gran mayoría de estudios. Debido al problema de pruebas múltiples, los indicadores locales son más propensos a identificar relaciones significativas, incluso cuando no existe asociación espacial significativa alguna. En este artículo los autores desarrollan una metodología que produce un índice local de Moran que proporciona tanto una estimación empírica de la escala espacial de la asociación así como una evaluación de la importancia del indicador para dicha escala. La idea clave es poner a prueba una serie de opciones posibles para la definición de la matriz de pesos espaciales (spatial weight matrix) del índice y luego tomar en cuenta las pruebas múltiples asociadas con el número de matrices de peso examinadas. A diferencia de métodos anteriores, el indicador local propuesto evita el uso de simulaciones para determinar la significancia estadística con pruebas múltiples. Para probar la validez del enfoque propuesto, se construyó un ejemplo numérico con el fin de evaluar el desempeño del nuevo índice y se llevó a cabo un estudio comparativo a partir de datos del centro de leucemia del estado de Nueva York. El índice desarrollado responde a la necesidad de definir las ponderaciones (pesos) empíricamente y la escala espacial. De esta forma el método propuesto supera limitaciones comúnmente halladas de muchas aplicaciones en las cuales los pesos son definidos exógenamente, con poca o ninguna atención a su definición o su implicancias. 局部空间统计量可用于度量和检验预定地理区域周围邻域的空间关联。大多数情况下仅采用单一尺度的分析而较少关注数据生成过程的尺度。而当其过程尺度无法确定时,局部统计量却可能检测出多个尺度。在这些案例中,对单个局部统计量统计显著性评估建立多重检验的修正是重要的,而这却鲜有实施。因此,即使存在无意义的空间关联时,局部统计也更可能识别出显著的相关性。 本文发展了一种基于局部Moran统计的方法,提供了空间尺度关联性的经验估计以及对该尺度下统计显著性的评估。其核心思想是测试统计权重矩阵的可能选择,然后考虑与权重矩阵检验数量数目相关的多重检验。与以往研究不同,该方法在多重检验情况中避免了采用模拟来确定统计显著性。为检验其有效性,采用了数值案例来评估其统计性能,并基于纽约州中部的血癌数据进行比较研究。该方法解决了权重和空间尺度确定经验估计的需求,通过验证也相应地解决了很多应用中的普遍弱点,即权重被定义成外生变量,而很少或根本没有考虑其定义或含义。     

Effect(s) of the “No-Same-Color-Touching” Constraint on the Join-Count Statistic: A Simulation Study

The join-count statistic is used to measure the tendency of polygons of a given map type to attract or repel polygons of the same or different map types. Yet in certain maps—for example, natural resources maps—it is often impossible for a polygon of a given type to touch another polygon of the same type. (This is the no-same-color or “No-Same-Type-Touching” (NSTT) constraint referred to in the title.) This violates an underlying assumption of the join-count statistic and may render its use to study certain spatial phenomena inappropriate—even for measuring spatial autocorrelation among polygons that are not the same type. This was explored using Monte Carlo simulation. For polygons of different types, it appears that results of the join-count statistic can be interpreted without any special consideration for the NSTT constraint provided there are a minimum of five to eight colors in the spatial system. For polygons of the same type, results can simply be ignored since it is known that no two polygons of the same type will touch.