Focal Location Quotients: Specification and Applications
|
|
- Merilyn Armstrong
- 6 years ago
- Views:
Transcription
1 Geographical Analysis (2012), Focal Location Quotients: Specification and Applications Robert G. Cromley, Dean M. Hanink Department of Geography, University of Connecticut, Storrs, CT, USA Location quotients (LQs) are commonly used descriptive statistics in spatial analysis. They are not directly affected by neighbor relationships, and their calculation ignores spatial information. This article computes LQs as focal, rather than strictly local, functions by incorporating the spatial structure of the observations in the reference group in their computation. Focal LQs, tested for significance by event-based Monte Carlo simulations, are applied both to North Carolina sudden infant death syndrome data for comparison with the G i *spatial statistic and to sectoral employment data in that state in a more typical context for LQ analysis. The significance tests show that the focal LQ is far more sensitive to the size of the denominator for spatially intensive data than is the G i *statistic. Introduction While the location quotient (LQ) can be used as a foundation for global measures (Mulligan and Schmidt 2005), it is perhaps the most commonly used local descriptive statistic in spatial analysis. Its use was widely established some time ago (Isard 1960), and it was widely employed in spatial research well before recent advances in geographic information science (GIScience) induced an emphasis on analysis at the local scale (Fotheringham 1997). Its use has been marked especially in the geographical analysis of economic activity; typical examples include the impact of geographical clustering on foreign investment (Fernhaber, Gilbert, and McDougall 2008), the location of manufacturing sales offices across the U.S. city system (Holmes 2005), the calculation of economic base multipliers (Riddington, Gibson, and Anderson 2006), and concentrations of U.S. automobile production (Carroll, Reid, and Smith 2008). Among noneconomic applications, LQs have been used to assess population distributions geographically (Brown and Chung 2006) and the spatial arrangement of health outcomes (Dira, Schaarschmidt, and Fayissa 2010). The usefulness of LQs as local descriptors is enhanced by their standardized form, which facilitates comparison not only to the reference set of observations in an aggregate but also across the individual units of observation. The mappability of LQs also facilitates cross-unit comparison and regionalization based on perceived patterns in their spatial distribution. While the utility of LQs in spatial analysis has been widely demonstrated, two of their limitations render them somewhat incomplete measures for spatial analysis. One is that they are Correspondence: Robert G. Cromley, Department of Geography, University of Connecticut, Campus Unit 4148, Storrs, CT robert.cromley@uconn.edu Submitted: June 27, Revised version accepted: April 3, doi: / x 2012 The Ohio State University 1
2 Geographical Analysis rarely subected to statistical significance testing. For example, a typical LQ with reference to a given sector s employment at observation (or location) i is a ratio of ratios. The ratio for the local unit of observation can be written as e i/e i, where e i is employment in the given sector at location i, and E i is total employment at location i; the ratio for the aggregate reference economy can be written as e/e where e and E are total employment in the given sector and overall employment in the reference economy, respectively. Then, LQi = ( ei Ei) ( e E). (1) Equal ratios provide the reference value of LQ i = 1; if LQ i < 1, then the ith place of observation has a relatively low level of employment in the sector in question, and if LQ i > 1, then it has a relatively high level. While the magnitude of a place s departure from the implicit null expectation that LQ i = 1 is indicated, whether that departure is statistically significant or simply occurs by reasonable chance is also of interest in attempting to understand whether locational patterns are meaningful in a functional as well as visual way. Important work about significance testing has been done with respect to computing confidence intervals for LQs using the delta method (Moineddin, Beyene, and Boyle 2003; Beyene and Moineddin 2005). The confidence limits provide probabilities concerning an individual LQ s difference from one. More recent work has been done by Dira, Schaarschmidt, and Fayissa (2010), who accommodate the negative comparisons among LQs in a sample (due to offsetting values among observations with respect to one) in their inferential treatment. The great maority of LQ analyses, however, employ the statistic purely for descriptive purposes. The second limitation affecting LQs in spatial analysis is that they are calculated without regard to neighborhood or other distance-related interaction effects among the units of observation. Useful explicitly spatial alternatives to LQs have been developed in the family of local indicator of spatial association (LISA) statistics, including the G statistics developed by Getis and Ord (1992) and the local Moran s I statistic (Anselin 1995). LISA statistics, like LQs, are calculated for individual units of observation. However, unlike LQs, LISA statistics incorporate the spatial structure of the units of observation in their computation and typically are subected to significance testing as a component of their interpretation. Regionalization or spatial cluster identification based upon LISA statistics can be evaluated not only by a visual criterion, as with LQs, but also by an inferential criterion (Jacquez 2008). Carroll, Reid, and Smith (2008) argue that LQs and G statistics are useful as complements in their analysis of automotive clusters in the United States, with the latter statistic providing the spatial interaction measure that is especially useful in regionalization. The purpose of this article is to provide a description of a focal LQ (FLQ) a relative ratio measure that incorporates the spatial structure of the observations in the reference group directly into its computation and that can be subected to significance testing in its interpretation. Like the LQ and the G statistic, the FLQ is a measure of spatial concentration. Unlike the standard LQ, however, the FLQ incorporates spatial information in its calculation and is a geographically weighted statistic. Further, unlike the G statistic, the computation of the FLQ uses geographically weighted aggregation for spatially intensive data. Therefore, the FLQ extends the foundations for measuring spatial concentration across areal units of observation. 2
3 Robert G. Cromley and Dean M. Hanink Focal Location Quotients Methodological foundation An emphasis on developing local statistical functions has led to somewhat of a local versus global dichotomy in spatial analysis. Calculated local statistics vary by location, and the output value is a function of more than one input location. In contrast, a global statistic is the same for all locations, and the output value is a function of all input locations (Fotheringham, Brunsdon, and Charlton 2002). Global and local statistics are based on the notion of an entire map pattern versus local map patterns (Openshaw 1991). Under this rubric, the LQ should be considered a local statistic because it describes the relationship at a location (an areal unit) within the context of a larger domain and can be mapped. However, a wider range of spatial definitions than this local/global dichotomy is necessary to distinguish between using information only within the areal unit itself versus also using information from areal units within a neighborhood. In defining cell-based, spatial functions for GIS operations used in cartographic modeling in general (Tomlin 1991) and ARCGRID in particular (ESRI 1994), a richer classification is used that encompasses local, focal, zonal, and global functions. Local functions compute an output value for each location based on only the input value(s) at that location, whereas global functions compute the output value at each location based on values at all other locations. Although it could, a global function is not required to produce the same value for all locations, so it differs from the concept of a whole-map pattern. The LQ should be a local function under this definition. The concept of the ARCGRID local function, however, differs from the local statistics used as indicators of spatial association (Getis and Ord 1992; Anselin 1995; Ord and Getis 1995) or for exploratory data analysis (Brunsdon, Fotheringham, and Charlton 2002) because no value for another location is used in their computations. The focal and zonal functions best match the LISA definitions of localized statistics. Focal functions calculate an output value for a location based on values from other locations within a specified neighborhood, and zonal functions compute a location s output value based on other locations within the same zone (or region). The former allows for a more continuous surface of output values, whereas the latter produces the same output value for every location within the same zone, producing a discrete, stepped surface of output values. Focal functions are most similar in nature to geographically weighted statistics as defined by Brunsdon, Fotheringham, and Charlton (2002). These scholars consider the FocalMedian function used in cartographic modeling to be a localized statistic. We prefer to use the nomenclature of cartographic modeling that is, focal LQ to distinguish the FLQ from the traditional LQ that is a local statistic. Because the denominator in the traditional LQ is a scalar, the output value is dependent only on input values from the location under investigation, and hence the LQ itself is a local operator. Furthermore, the numerator is a ratio that also is a local operator. In addition to being a ratio of ratios, the LQ also can be written as LQi = ei [ Ei( e E)]. (2) Here, the numerator is the observed number of events, and the denominator is the expected number of events, for the ith observational unit. As a measure of concentration, values greater than one signify that more of an event is occurring, and for values less than one, less than expected is occurring. Geographically weighted statistics commonly are a linear combination of variate values: LS = w u, (3) i i 3
4 Geographical Analysis where w i is a standardized geographic weight, and u is the th value of the variate of interest. Weights may be based on some type of kernel density functions, such as a box kernel, a Gaussian kernel, or a bisquare kernel (see Fotheringham, Brunsdon, and Charlton 2002, pp , for specific definitions). When observations correspond to areal units, the variate values can be either spatially extensive (e.g., counts) or spatially intensive (e.g., rates, proportions, and ratios; Goodchild and Lam 1980). A geographically weighted statistic for count data uses the form given by equation (3). However, for spatially intensive data in which z = x /y, one can either geographically weight the z values, as in equation (3), or geographically weight the numerators (x ) separately from the denominators (y ) of the intensive values before their ratio is taken, such as. (4) LS wx wy i = ( i ) ( i ) The latter approach treats the statistic as a geographically weighted aggregation of the individual obects within each original areal unit that gives rise to the spatially intensive value rather than a direct measurement of the areal units themselves such as perimeter. Because the LQ is defined as a ratio of ratios, it is a ratio of two spatially intensive values, and this alternative weighting approach is used here to calculate FLQs as ( ) (5) FLQ w e w E e E i = ( i i ) The attributes of any areal aggregation unit are themselves based on geographically weighted summaries of individual data contained within an areal unit. The weight is equal to one for each individual located within the boundary of that areal unit and is equal to zero for each individual located outside the boundary of that areal unit. The focal approach extends the aggregation to locations outside an initial areal unit. The LQ is normally calculated for discrete areal units because most economic data collected by governments are reported only at higher levels of aggregation due to confidentiality restrictions. However, data for individual firm locations are available through private vendors, and no theoretical reason exists for precluding an FLQ at any location. This conceptualization mirrors Rushton and Lolonis (1996) conception of health rates as a continuous spatial distribution rather than as one based on administrative units, as well as the generation of regression parameters and summary statistics (Brunsdon, Fotheringham, and Charlton 2002) as a continuous surface. The FLQ is somewhat similar to the G i * statistic (Getis and Ord 1992) when using spatially intensive data but differs from it in certain important respects. The G i * statistic, unlike the FLQ, does not use geographically weighted aggregation and is defined as * = (6) G w z z i i where z is the variate value of interest. For spatially intensive data, i = ( i( ) ) ( ) G * w x y x y. (7) The G i * statistic is a ratio of the sum of weighted spatially intensive values to the sum of the respective unweighted values, which Ord and Getis (2001) note is simply a weighted moving average. A geographically weighted aggregation of spatially intensive values sums the weighted 4
5 Robert G. Cromley and Dean M. Hanink Focal Location Quotients numerators and sums the weighted denominators that form the spatially intensive values separately before taking the ratios, as follows: (( wx i wy x y ) ( i )) (( ) ( )), (8) which is the same as the FLQ. Because spatially intensive data for an areal unit are aggregated separately by numerator and by denominator before a ratio is calculated, aggregating geographically weighted numerators and denominators separately before ratios are calculated in computing statistics that encompass data beyond areal unit boundaries is more natural. This order of aggregation also will have an effect on significance testing. Although FLQs are easy to define as descriptive statistics, determining if any value is significantly different from one in order to identify meaningful hot (high FLQ) or cold (low FLQ) spots also is important. Recently, confidence intervals for traditional LQs have been established for this purpose (see Moineddin, Beyene, and Boyle 2003; Dira, Schaarschmidt, and Fayissa 2010). Significance tests for FLQs are produced empirically in this study using a technique developed by Rushton and Lolonis (1996) to determine the probability that an observed rate is significant. In the Rushton Lolonis method, a distribution of test statistics is created by 1,000 simulations in which each individual event is given a chance of acquiring the state under investigation. A random number is generated from a uniform distribution in the range 1 1,000. The rate observed for the whole region (the denominator of the FLQ), expressed as N per 1,000, is used to determine whether the state occurs for an individual event. The number of events is equal to the total number of employees in the case of sectoral employment or to the total number of births in the case of sudden infant death syndrome (SIDS). Each event has a geocoded areal address for aggregated data, such as county, or a geocoded (x,y) address for disaggregated data. If the random number is in the range 1 N, the individual event is given the simulated state; otherwise, the event does not have that state. This step is repeated for every individual event in the study region to produce a simulated pattern of FLQ. One thousand such simulations were executed, and 1,000 different FLQs for each location were rank ordered in a distribution. The observed FLQ for each location was compared with the simulated distribution of FLQs for that location to determine its significance level. Like G i *, the FLQ is scale invariant in the calculation of the statistic; however, its significance level changes as the size of the population and/or number of cases changes. Also, for spatially intensive data in which both the numerator and denominator are scaled by the same constant, the z variate values used to calculate the G i * statistic do not change even when the numerators and denominators change. Unlike the G i * statistic (whose significance levels are based on a classical distribution), the significance level of the FLQ based on simulations should be tighter for the data sets with much larger underlying populations (denominators) and/or cases (numerators). Empirical examples A shapefile for North Carolina counties was downloaded from the U.S. Bureau of the Census website for mapping and distance calculations. The 100 counties of North Carolina also are regionalized into the Mountains, the Piedmont, and the Coastal Plain regions of the state (Fig. 1). The original geographic coordinates were proected into North Carolina state plane coordinates using the WGS84 datum. For mapping purposes, the following statistical significance classes are used: less than 2.5%, 2.5% to 12.5%, 12.5% to 87.5%, 87.5% to 97.5, and greater than 97.5%. 5
6 Geographical Analysis Figure 1. North Carolina counties and regions. At the 95% significance level, those less than 2.5% are significantly below expectation (ª1.0), and those greater than 2.5% were significantly above expectation. The Cressie and Chan (1989) SIDS data set for North Carolina and North Carolina employment data for December 2007 constitute the illustrative data sets. All data are aggregated to the county level for the 100 counties of North Carolina. The SIDS rates for the period are examined first. Overall, 836 SIDS cases occurred for 422,392 births, a rate of 2 per 1,000. A single Monte Carlo simulation requires 422,392 events one for every birth. For each birth, a random number is generated in the range 1 1,000, and the birth is given the simulated state (a SIDS death occurs) whenever the number is less than or equal to two; otherwise, the birth does not have that state. Because births are spatially post-stratified by county, the simulated deaths among these births also are post-stratified by county. Each simulation generates a distribution of SIDS cases by county that can be used to calculate an LQ and a FLQ statistic for each county. A distribution of LQs and FLQs for each county can be created by executing 1,000 simulations. Fig. 2a presents the actual county-based LQs for SIDS, and Fig. 2b presents the associated significance level based on Monte Carlo simulation. Six counties, located mainly in the southern tier, are significantly higher, and 13 counties dispersed across the state are significantly lower than expected. The FLQs can be calculated using a box kernel with a 53-km (33-mile) bandwidth. This kernel is the same as originally used by Getis and Ord (1992) to compute their G i * statistic for the same data set. Distances used are calculated between county centroids. Fig. 3a presents the actual FLQs for SIDS, and Fig. 3b presents the associated significance levels based on Monte Carlo simulation. Now 12 counties have significantly high FLQs, and 11 counties, clustered mainly in a north-to-south band in the Piedmont region, have significantly low FLQs. The G i * map (Fig. 3c) shows a similar pattern of significance, although it contains fewer highly significant locations than the FLQs map. Because the significance tests for the FLQ are based on many more obects than for the G i * statistic (the number of events within the neighborhood of a given areal unit rather than ust the number of areal units within its neighborhood), deviations from the expected value are more likely to be significant. Next, the spatially continuous version of the FLQ (its value is not constant within an areal unit) is illustrated using the aggregated SIDS data because no spatial information about individual births and deaths is available. To compute the continuous FLQ surface, a Gaussian kernel with a 53-km (33-mile) bandwidth was applied to a regular square grid with 1-mile spacing between points (approximately 49,000 locations). Distances were calculated between each point 6
7 Robert G. Cromley and Dean M. Hanink Focal Location Quotients Figure 2. LQs and significance levels for North Carolina SIDS cases, (a) LQs for SIDS cases. (b) Significance level for LQs. and the county centroids. Fig. 4a presents a map of the continuous FLQ surface, and Fig. 4b portrays the associated significance levels (the significance was not adusted for multiple testing). The pattern of significance for the continuous surface conforms to that of the discrete rendition. LQs are most commonly used in analyzing economic data, and sectoral employment frequently is the variable of interest. The following analysis furnishes two such applications with employment data investigated for the finance and real estate sector and for the manufacturing sector (Bureau of Labor Statistics 2011). Only the discrete version of the FLQ is calculated and mapped because no individual firm data were available and the mechanics of the continuous version is outlined in the previous SIDS example. From a total employed labor force of 5,460,841, 209,117 individuals are employed in the finance and real estate sector, a rate of 38 per 1,000. A single Monte Carlo simulation in this instance has 5,460,841 events, and each event is assigned an employment case in the finance and real estate sector whenever the random number is less than or equal to 38. Financial and real estate services in North Carolina are highly concentrated in Mecklenburg County, the location of Charlotte, a maor regional banking and insurance center. That county is the only one in the state with an LQ greater than two (Fig. 5a). Besides Mecklenburg, other counties with LQs greater than one include those containing the cities of Winston-Salem, Greensboro, Durham, and Raleigh. The discrete FLQs are greater than one for counties in a band running through the Piedmont region from Winston-Salem and Greensboro south to Charlotte. The pattern of significance (Fig. 5b) exactly matches the pattern of FLQs: those counties with FLQs less than one are significantly below and those with FLQs 7
8 Geographical Analysis Figure 3. A comparison of the FLQ for SIDS cases at a 53-km (33-mile) distance and its significance level, and the significance level of the G i * Statistic at the same distance for North Carolina SIDS cases, (a) FLQs. (b) Significance level for FLQs. (c) Significance level for the G i * statistic. greater than one are significantly above the expected value. The pattern for the G i * statistic is somewhat different. No county is significantly below, and only Mecklenburg and four of its neighboring counties and the counties containing Winston-Salem and Greensboro are significantly above at the 5% level (Fig. 5c). From the total employed labor force, 559,913 individuals are employed in the manufacturing sector, a rate of 103 per 1,000. The Monte Carlo simulation for manufacturing uses the same number of events as the finance and real estate sector simulation, but a manufacturing case occurs 8
9 Robert G. Cromley and Dean M. Hanink Focal Location Quotients Figure 4. The pattern of the continuous FLQs and significance levels for SIDS cases at a 53-km (33-mile) bandwidth, for North Carolina SIDS cases, (a) Continuous FLQs. (b) Continuous FLQ significance levels. whenever the generated random number is less than or equal to 103. Manufacturing employment is more widely distributed in North Carolina than finance and real estate employment, with the lowest concentrations in the western mountain counties and the eastern Coastal Plain counties (Fig. 6a). The pattern of FLQs again effectively corresponds to their 5% significance level; with one exception, all counties with a FLQ value below one are significantly lower and all counties above one are significantly higher than expected (Fig. 6b). The G i * pattern is significantly higher in the Piedmont region (Fig. 6c), while some areas in the west and coastal tidewater region are significantly lower. To examine the impact of differences in the number of cases and population size, the number of SIDS cases and the number of live births for each county were first multiplied by a factor of 6 and then by a factor of 12; these multiplicative factors increase the population size without changing the SIDS mortality rate for the entire region. As the size proportionally increases, the number of extreme FLQs increases. For the original values, only 22 counties have extreme values. The number of extreme values increases to 58 with the sixfold increase in SIDS cases and live births, and to 71 with the 12-fold increase. All FLQs will be extreme (in either the upper or the lower tail of the simulated distribution) when the population reaches a certain size, unless a calculated value approximates the expected value of one. Finally, the overall SIDS mortality rate increases from 2 per 1,000 to 20 per 1,000 when the numerator increases 10-fold for every county; this change does not alter any FLQ values. With this latter 9
10 Geographical Analysis Figure 5. Comparison of the pattern of LQs and significance levels for the FLQ and G i * statistic for the financial and real estate sector employment in North Carolina, December (a) LQs. (b) Significance level for the FLQs. (c) Significance level for the G i * statistic. increase, the number of extreme FLQs increases from 22 to 65. Again, increasing the size of the number of cases in the simulated distribution increases the likelihood that any variation is statistically significant. Therefore, the number of extreme values is a function of the size of the population as well as of the rate of the event within the population for FLQs, which is not the case for the G i * statistic. Summary and conclusions The LQ is a descriptive statistic of concentration widely used in economic geography but easily applied to a variety of spatial data. Its utility as a concentration measure can be further enhanced 10
11 Robert G. Cromley and Dean M. Hanink Focal Location Quotients Figure 6. Comparison of the pattern of LQs and significance levels for the FLQs and G i * statistic for manufacturing sector employment in North Carolina, December (a) LQs. (b) Significance level for the FLQs. (c) Significance level for the G i * statistic. if its geographic domain is made variable. The concept of FLQs developed here is an attempt to implement a more spatially versatile measure that can detect levels of concentration in event outcomes across a range of geographic scales. The FLQ is an alternative to other local concentration measures, such as those derived from Moran s I and the G statistics, when using spatially intensive variate values associated with aggregated areal data. In this situation, the FLQ is shown to be a geographically weighted aggregation counterpart to the G i * statistic. In addition to calculating FLQs for discrete areal units, this article shows how to generate a continuous surface of FLQs based on individual point data. The purpose of the continuous 11
12 Geographical Analysis measure is to reflect a distribution unencumbered by an arbitrary partitioning of space. In the absence of individual data, spatially aggregated data could be used, which is illustrated with Cressie and Chan s North Carolina SIDS example. The discrete and continuous versions of the FLQ can be used by health care policy makers to examine disparities in health status over a range of bandwidths to determine the scale at which the disparities occur. Significance tests were developed using a Monte Carlo simulation of alternative data distributions. In empirical tests involving SIDS data and economic employment sector data for the state of North Carolina, extreme values are shown to be a function of the overall size of the population and of the rate of the event within the entire population. When using spatially intensive data, an analyst should be aware of the overall population size and may want to investigate spatial clusters using FLQs as an alternative to existing measures. Spatially intensive variate values for areal obect data should be handled differently than variate values for field data, and the aggregation aspect of the former data should be recognized. Spatially intensive variate values based on larger denominators are more significant than ones based on smaller denominators; this size effect is lost when the data are viewed as directly measured variate values. FLQs should be found especially useful in providing an explicitly spatial component to those types of analyses that commonly use traditional LQs as building blocks for regionalization. In addition, FLQs also can eliminate the necessity of using traditional LQs in tandem with LISA statistics in assessing univariate spatial distributions. References Anselin, L. (1995). Local Indicators of Spatial Association-LISA. Geographical Analysis 27(2), Beyene, J., and R. Moineddin. (2005). Methods for Confidence Interval Estimation of a Ratio Parameter with Application to Location Quotients. BMC Medical Research Methodology 5, 32. Brown, L., and S.-Y. Chung. (2006). Spatial Segregation, Segregation Indices and the Geographical Perspective. Population, Space and Place 12(2), Brunsdon, C., A. S. Fotheringham, and M. Charlton. (2002). Geographically Weighted Summary Statistics A Framework for Localised Exploratory Data Analysis. Computers, Environment and Urban Systems 26(6), Bureau of Labor Statistics. (2011). Local Area Unemployment Statistics. Available at (accessed on 15 January 2011). Carroll, M., N. Reid, and B. Smith. (2008). Location Quotients Versus Spatial Autocorrelation in Identifying Potential Cluster Regions. Annals of Regional Science 42(2), Cressie, N., and N. Chan. (1989). Spatial Modelling of Regional Variables. Journal of the American Statistical Association 84(406), Dira, G., F. Schaarschmidt, and B. Fayissa. (2010). Inferences for Selected Location Quotients with Applications to Health Outcomes. Geographical Analysis 42(3), ESRI. (1994). Cell-Based Modeling with GRID. Redlands, CA: Environmental Systems Research Institute. Fernhaber, S., B. Gilbert, and P. McDougall. (2008). International Entrepreneurship and Geographic Location: An Empirical Examination of New Venture Internationalization. Journal of International Business Studies 39(2), Fotheringham, A. S. (1997). Trends in Quantitative Methods I: Stressing the Local. Progress in Human Geography 21(1), Fotheringham, A. S., C. Brunsdon, and M. Charlton. (2002). Geographically Weighted Regression: The Analysis of Spatially Varying Relationships. West Sussex, U.K.: Wiley. Getis, A., and J. Ord. (1992). The Analysis of Spatial Association by Use of Distance Statistics. Geographical Analysis 24(3),
13 Robert G. Cromley and Dean M. Hanink Focal Location Quotients Goodchild, M., and N. Lam. (1980). Areal Interpolation: Variant of the Traditional Spatial Problem. Geo-Processing 1, Holmes, T. (2005). The Location of Sales Offices and the Attraction of Cities. Journal of Political Economy 113(3), Isard, W. (1960). Methods of Regional Analysis. Cambridge: MIT Press. Jacquez, G. (2008). Spatial Cluster Analysis. In The Handbook of Geographic Information Science, , edited by S. Fotheringham and J. Wilson. Oxford: Blackwell. Moineddin, R., J. Beyene, and E. Boyle. (2003). On the Location Quotient Confidence Interval. Geographical Analysis 35(3), Mulligan, G., and C. Schmidt. (2005). A Note on Localization and Specialization. Growth and Change 36(4), Openshaw, S. (1991). Developing Spatial Analysis Methods for GIS. In Geographical Information Systems: Principles and Applications, , edited by D. Maguire, M. Goodchild and D. Rhind. London: Longman. Ord, J., and A. Getis. (1995). Local Spatial Autocorrelation Statistics: Distributional Issues and an Application. Geographical Analysis 27(4), Ord, J., and A. Getis. (2001). Testing for Local Spatial Autocorrelation in the Presence of Global Autocorrelation. Journal of Regional Science 41(3), Riddington, G., H. Gibson, and J. Anderson. (2006). Comparison of Gravity Model, Survey and Location Quotient-Based Local Area Tables and Multipliers. Regional Studies 40(9), Rushton, G., and P. Lolonis. (1996). Exploratory Spatial Analysis of Birth Defect Rates in an Urban Population. Statistics in Medicine 15(7 9), Tomlin, C. D. (1991). Cartographic Modelling. In Geographical Information Systems: Principles and Applications, , edited by D. Maguire, M. Goodchild and D. Rhind. London: Longman. 13
GIS Spatial Statistics for Public Opinion Survey Response Rates
GIS Spatial Statistics for Public Opinion Survey Response Rates July 22, 2015 Timothy Michalowski Senior Statistical GIS Analyst Abt SRBI - New York, NY t.michalowski@srbi.com www.srbi.com Introduction
More informationLuc Anselin Spatial Analysis Laboratory Dept. Agricultural and Consumer Economics University of Illinois, Urbana-Champaign
GIS and Spatial Analysis Luc Anselin Spatial Analysis Laboratory Dept. Agricultural and Consumer Economics University of Illinois, Urbana-Champaign http://sal.agecon.uiuc.edu Outline GIS and Spatial Analysis
More informationSpatial Analysis 1. Introduction
Spatial Analysis 1 Introduction Geo-referenced Data (not any data) x, y coordinates (e.g., lat., long.) ------------------------------------------------------ - Table of Data: Obs. # x y Variables -------------------------------------
More informationPOPULAR CARTOGRAPHIC AREAL INTERPOLATION METHODS VIEWED FROM A GEOSTATISTICAL PERSPECTIVE
CO-282 POPULAR CARTOGRAPHIC AREAL INTERPOLATION METHODS VIEWED FROM A GEOSTATISTICAL PERSPECTIVE KYRIAKIDIS P. University of California Santa Barbara, MYTILENE, GREECE ABSTRACT Cartographic areal interpolation
More information1Department of Demography and Organization Studies, University of Texas at San Antonio, One UTSA Circle, San Antonio, TX
Well, it depends on where you're born: A practical application of geographically weighted regression to the study of infant mortality in the U.S. P. Johnelle Sparks and Corey S. Sparks 1 Introduction Infant
More informationThis report details analyses and methodologies used to examine and visualize the spatial and nonspatial
Analysis Summary: Acute Myocardial Infarction and Social Determinants of Health Acute Myocardial Infarction Study Summary March 2014 Project Summary :: Purpose This report details analyses and methodologies
More informationSPACE Workshop NSF NCGIA CSISS UCGIS SDSU. Aldstadt, Getis, Jankowski, Rey, Weeks SDSU F. Goodchild, M. Goodchild, Janelle, Rebich UCSB
SPACE Workshop NSF NCGIA CSISS UCGIS SDSU Aldstadt, Getis, Jankowski, Rey, Weeks SDSU F. Goodchild, M. Goodchild, Janelle, Rebich UCSB August 2-8, 2004 San Diego State University Some Examples of Spatial
More informationIdentification of Economic Clusters Using ArcGIS Spatial Statistics. Joseph Frizado Bruce Smith Michael Carroll
Identification of Economic Clusters Using ArcGIS Spatial Statistics Joseph Frizado Bruce Smith Michael Carroll ABSTRACT Geographic proximity (co-location) is necessary for potential clustering activity.
More informationExploratory Spatial Data Analysis (ESDA)
Exploratory Spatial Data Analysis (ESDA) VANGHR s method of ESDA follows a typical geospatial framework of selecting variables, exploring spatial patterns, and regression analysis. The primary software
More informationGIS SPATIAL ANALYSIS OF UNIVERSITY OF NEBRASKA AT KEARNEY ALUMNI COHORTS,
University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln Great Plains Research: A Journal of Natural and Social Sciences Great Plains Studies, Center for 2010 GIS SPATIAL ANALYSIS
More informationMapping and Analysis for Spatial Social Science
Mapping and Analysis for Spatial Social Science Luc Anselin Spatial Analysis Laboratory Dept. Agricultural and Consumer Economics University of Illinois, Urbana-Champaign http://sal.agecon.uiuc.edu Outline
More informationFinding Hot Spots in ArcGIS Online: Minimizing the Subjectivity of Visual Analysis. Nicholas M. Giner Esri Parrish S.
Finding Hot Spots in ArcGIS Online: Minimizing the Subjectivity of Visual Analysis Nicholas M. Giner Esri Parrish S. Henderson FBI Agenda The subjectivity of maps What is Hot Spot Analysis? Why do Hot
More informationThis lab exercise will try to answer these questions using spatial statistics in a geographic information system (GIS) context.
by Introduction Problem Do the patterns of forest fires change over time? Do forest fires occur in clusters, and do the clusters change over time? Is this information useful in fighting forest fires? This
More informationSpatial Trends of unpaid caregiving in Ireland
Spatial Trends of unpaid caregiving in Ireland Stamatis Kalogirou 1,*, Ronan Foley 2 1. NCG Affiliate, Thoukididi 20, Drama, 66100, Greece; Tel: +30 6977 476776; Email: skalogirou@gmail.com; Web: http://www.gisc.gr.
More informationTracey Farrigan Research Geographer USDA-Economic Research Service
Rural Poverty Symposium Federal Reserve Bank of Atlanta December 2-3, 2013 Tracey Farrigan Research Geographer USDA-Economic Research Service Justification Increasing demand for sub-county analysis Policy
More informationESRI 2008 Health GIS Conference
ESRI 2008 Health GIS Conference An Exploration of Geographically Weighted Regression on Spatial Non- Stationarity and Principal Component Extraction of Determinative Information from Robust Datasets A
More informationNeighborhood social characteristics and chronic disease outcomes: does the geographic scale of neighborhood matter? Malia Jones
Neighborhood social characteristics and chronic disease outcomes: does the geographic scale of neighborhood matter? Malia Jones Prepared for consideration for PAA 2013 Short Abstract Empirical research
More informationLocal Spatial Autocorrelation Clusters
Local Spatial Autocorrelation Clusters Luc Anselin http://spatial.uchicago.edu LISA principle local Moran local G statistics issues and interpretation LISA Principle Clustering vs Clusters global spatial
More informationSpatial Analysis I. Spatial data analysis Spatial analysis and inference
Spatial Analysis I Spatial data analysis Spatial analysis and inference Roadmap Outline: What is spatial analysis? Spatial Joins Step 1: Analysis of attributes Step 2: Preparing for analyses: working with
More informationChapter 6 Spatial Analysis
6.1 Introduction Chapter 6 Spatial Analysis Spatial analysis, in a narrow sense, is a set of mathematical (and usually statistical) tools used to find order and patterns in spatial phenomena. Spatial patterns
More informationContext-dependent spatial analysis: A role for GIS?
J Geograph Syst (2000) 2:71±76 ( Springer-Verlag 2000 Context-dependent spatial analysis: A role for GIS? A. Stewart Fotheringham Department of Geography, University of Newcastle, Newcastle-upon-Tyne NE1
More informationWhere to Invest Affordable Housing Dollars in Polk County?: A Spatial Analysis of Opportunity Areas
Resilient Neighborhoods Technical Reports and White Papers Resilient Neighborhoods Initiative 6-2014 Where to Invest Affordable Housing Dollars in Polk County?: A Spatial Analysis of Opportunity Areas
More informationComparison of spatial methods for measuring road accident hotspots : a case study of London
Journal of Maps ISSN: (Print) 1744-5647 (Online) Journal homepage: http://www.tandfonline.com/loi/tjom20 Comparison of spatial methods for measuring road accident hotspots : a case study of London Tessa
More informationUsing AMOEBA to Create a Spatial Weights Matrix and Identify Spatial Clusters, and a Comparison to Other Clustering Algorithms
Using AMOEBA to Create a Spatial Weights Matrix and Identify Spatial Clusters, and a Comparison to Other Clustering Algorithms Arthur Getis* and Jared Aldstadt** *San Diego State University **SDSU/UCSB
More informationSASI Spatial Analysis SSC Meeting Aug 2010 Habitat Document 5
OBJECTIVES The objectives of the SASI Spatial Analysis were to (1) explore the spatial structure of the asymptotic area swept (z ), (2) define clusters of high and low z for each gear type, (3) determine
More informationGeography Department. Summer transition work
Geography Department Summer transition work An essential part of studying Geography post-16 is becoming familiar with statistical testing for fieldwork. Please complete the following activities over summer
More informationDepartment of Geography, University of Connecticut, Storrs, CT, USA. Online publication date: 28 March 2011 PLEASE SCROLL DOWN FOR ARTICLE
This article was downloaded by: [University of Connecticut] On: 28 March 2011 Access details: Access Details: [subscription number 784375807] Publisher Taylor & Francis Informa Ltd Registered in England
More informationTexas A&M University
Texas A&M University CVEN 658 Civil Engineering Applications of GIS Hotspot Analysis of Highway Accident Spatial Pattern Based on Network Spatial Weights Instructor: Dr. Francisco Olivera Author: Zachry
More informationDaniel Fuller Lise Gauvin Yan Kestens
Examining the spatial distribution and relationship between support for policies aimed at active living in transportation and transportation behavior Daniel Fuller Lise Gauvin Yan Kestens Introduction
More informationThe Use of Spatial Weights Matrices and the Effect of Geometry and Geographical Scale
The Use of Spatial Weights Matrices and the Effect of Geometry and Geographical Scale António Manuel RODRIGUES 1, José António TENEDÓRIO 2 1 Research fellow, e-geo Centre for Geography and Regional Planning,
More informationLOCATIONAL PREFERENCES OF FDI FIRMS IN TURKEY
LOCATIONAL PREFERENCES OF FDI FIRMS IN TURKEY Prof. Dr. Lale BERKÖZ Assist. Prof. Dr.S. SenceTÜRK I.T.U. Faculty of Architecture Istanbul/TURKEY E-mail: lberkoz@itu.edu.tr INTRODUCTION Foreign direct investment
More informationExploratory Spatial Data Analysis of Regional Economic Disparities in Beijing during the Preparation Period of the 2008 Olympic Games
Exploratory Spatial Data Analysis of Regional Economic Disparities in Beijing during the Preparation Period of the 2008 Olympic Games Xiaoyi Ma, Tao Pei Thursday, May 27, 2010 The State Key Laboratory
More informationBROOKINGS May
Appendix 1. Technical Methodology This study combines detailed data on transit systems, demographics, and employment to determine the accessibility of jobs via transit within and across the country s 100
More informationSpatial and Temporal Geovisualisation and Data Mining of Road Traffic Accidents in Christchurch, New Zealand
166 Spatial and Temporal Geovisualisation and Data Mining of Road Traffic Accidents in Christchurch, New Zealand Clive E. SABEL and Phil BARTIE Abstract This paper outlines the development of a method
More informationEXPLORATORY SPATIAL DATA ANALYSIS OF BUILDING ENERGY IN URBAN ENVIRONMENTS. Food Machinery and Equipment, Tianjin , China
EXPLORATORY SPATIAL DATA ANALYSIS OF BUILDING ENERGY IN URBAN ENVIRONMENTS Wei Tian 1,2, Lai Wei 1,2, Pieter de Wilde 3, Song Yang 1,2, QingXin Meng 1 1 College of Mechanical Engineering, Tianjin University
More informationCapital, Institutions and Urban Growth Systems
Capital, Institutions and Urban Growth Systems Robert Huggins Centre for Economic Geography, School of Planning and Geography, Cardiff University Divergent Cities Conference, University of Cambridge, Cambridge
More informationFormalization of GIS functionality
Formalization of GIS functionality Over the past four decades humans have invested significantly in the construction of tools for handling digital representations of spaces and their contents. These include
More informationSpatial Clusters of Rates
Spatial Clusters of Rates Luc Anselin http://spatial.uchicago.edu concepts EBI local Moran scan statistics Concepts Rates as Risk from counts (spatially extensive) to rates (spatially intensive) rate =
More informationIntroduction to Spatial Statistics and Modeling for Regional Analysis
Introduction to Spatial Statistics and Modeling for Regional Analysis Dr. Xinyue Ye, Assistant Professor Center for Regional Development (Department of Commerce EDA University Center) & School of Earth,
More informationGeographically weighted regression approach for origin-destination flows
Geographically weighted regression approach for origin-destination flows Kazuki Tamesue 1 and Morito Tsutsumi 2 1 Graduate School of Information and Engineering, University of Tsukuba 1-1-1 Tennodai, Tsukuba,
More informationLocational Error Impacts on Local Spatial Autocorrelation Indices: A Syracuse Soil Sample Pb-level Data Case Study
Locational Error Impacts on Local Spatial Autocorrelation Indices: A Syracuse Soil Sample Pb-level Data Case Study Daniel A. Griffith *1, Yongwan Chun 1, and Monghyeon Lee 1 1 University of Texas at Dallas,
More informationEconomic Geography of the Long Island Region
Geography of Data Economic Geography of the Long Island Region Copyright 2011 AFG 1 The geography of economic activity requires: - the gathering of spatial data - the location of data geographically -
More informationSpatial Variation in Hospitalizations for Cardiometabolic Ambulatory Care Sensitive Conditions Across Canada
Spatial Variation in Hospitalizations for Cardiometabolic Ambulatory Care Sensitive Conditions Across Canada CRDCN Conference November 14, 2017 Martin Cooke Alana Maltby Sarah Singh Piotr Wilk Today s
More informationAnalysis of Bank Branches in the Greater Los Angeles Region
Analysis of Bank Branches in the Greater Los Angeles Region Brian Moore Introduction The Community Reinvestment Act, passed by Congress in 1977, was written to address redlining by financial institutions.
More informationOutline. 15. Descriptive Summary, Design, and Inference. Descriptive summaries. Data mining. The centroid
Outline 15. Descriptive Summary, Design, and Inference Geographic Information Systems and Science SECOND EDITION Paul A. Longley, Michael F. Goodchild, David J. Maguire, David W. Rhind 2005 John Wiley
More informationSPATIAL ANALYSIS. Transformation. Cartogram Central. 14 & 15. Query, Measurement, Transformation, Descriptive Summary, Design, and Inference
14 & 15. Query, Measurement, Transformation, Descriptive Summary, Design, and Inference Geographic Information Systems and Science SECOND EDITION Paul A. Longley, Michael F. Goodchild, David J. Maguire,
More informationOutline ESDA. Exploratory Spatial Data Analysis ESDA. Luc Anselin
Exploratory Spatial Data Analysis ESDA Luc Anselin University of Illinois, Urbana-Champaign http://www.spacestat.com Outline ESDA Exploring Spatial Patterns Global Spatial Autocorrelation Local Spatial
More informationGIS and Spatial Statistics: One World View or Two? Michael F. Goodchild University of California Santa Barbara
GIS and Spatial Statistics: One World View or Two? Michael F. Goodchild University of California Santa Barbara Location as attribute The data table Census summary table What value is location as an explanatory
More informationCSISS Tools and Spatial Analysis Software
CSISS Tools and Spatial Analysis Software June 5, 2006 Stephen A. Matthews Associate Professor of Sociology & Anthropology, Geography and Demography Director of the Geographic Information Analysis Core
More informationA 3D GEOVISUALIZATION APPROACH TO CRIME MAPPING
A 3D GEOVISUALIZATION APPROACH TO CRIME MAPPING Markus Wolff, Hartmut Asche 3D-Geoinformation Research Group Department of geography University of Potsdam Markus.Wolff@uni-potsdam.de, gislab@uni-potsdam.de
More informationInclusion of Non-Street Addresses in Cancer Cluster Analysis
Inclusion of Non-Street Addresses in Cancer Cluster Analysis Sue-Min Lai, Zhimin Shen, Darin Banks Kansas Cancer Registry University of Kansas Medical Center KCR (Kansas Cancer Registry) KCR: population-based
More informationGuilty of committing ecological fallacy?
GIS: Guilty of committing ecological fallacy? David W. Wong Professor Geography and GeoInformation Science George Mason University dwong2@gmu.edu Ecological Fallacy (EF) Many slightly different definitions
More informationHUMAN CAPITAL CATEGORY INTERACTION PATTERN TO ECONOMIC GROWTH OF ASEAN MEMBER COUNTRIES IN 2015 BY USING GEODA GEO-INFORMATION TECHNOLOGY DATA
International Journal of Civil Engineering and Technology (IJCIET) Volume 8, Issue 11, November 2017, pp. 889 900, Article ID: IJCIET_08_11_089 Available online at http://http://www.iaeme.com/ijciet/issues.asp?jtype=ijciet&vtype=8&itype=11
More informationApplications of GIS in Health Research. West Nile virus
Applications of GIS in Health Research West Nile virus Outline Part 1. Applications of GIS in Health research or spatial epidemiology Disease Mapping Cluster Detection Spatial Exposure Assessment Assessment
More informationGeographical General Regression Neural Network (GGRNN) Tool For Geographically Weighted Regression Analysis
Geographical General Regression Neural Network (GGRNN) Tool For Geographically Weighted Regression Analysis Muhammad Irfan, Aleksandra Koj, Hywel R. Thomas, Majid Sedighi Geoenvironmental Research Centre,
More informationIntroduction to Geographic Information Systems
Geog 58 Introduction to Geographic Information Systems, Fall, 2003 Page 1/8 Geography 58 Introduction to Geographic Information Systems Instructor: Lecture Hours: Lab Hours: X-period: Office Hours: Classroom:
More information2012 State of the Region Address. Michael C. Carroll, Ph.D. Center for Regional Development Bowling Green State University
2012 State of the Region Address Michael C. Carroll, Ph.D. Center for Regional Development Bowling Green State University Outline How we have changed Changing employment trends Temporal view of unemployment
More informationExploring Digital Welfare data using GeoTools and Grids
Exploring Digital Welfare data using GeoTools and Grids Hodkinson, S.N., Turner, A.G.D. School of Geography, University of Leeds June 20, 2014 Summary As part of the Digital Welfare project [1] a Java
More informationTemporal vs. Spatial Data
Temporal vs. Spatial Data Temporal 1 dimensional Units: day, week, month Lag: t, t-1, t-2 Durbin-Watson Spatial 2-3 dimensional Units: county, mile, region Lag: near neighbor, networks (?) Moran s I Differencing
More informationSpatial nonstationarity and autoregressive models
Environment and Planning A 1998, volume 30, pages 957-973 Spatial nonstationarity and autoregressive models C Brunsdon Department of Town and Country Planning, University of Newcastle, Newcastle upon Tyne
More informationMeasuring community health outcomes: New approaches for public health services research
Research Brief March 2015 Measuring community health outcomes: New approaches for public health services research P ublic Health agencies are increasingly asked to do more with less. Tough economic times
More informationGIS Test Drive What a Geographic Information System Is and What it Can Do. Alison Davis-Holland
GIS Test Drive What a Geographic Information System Is and What it Can Do Alison Davis-Holland adavisholland@gmail.com WHO AM I? Geospatial Analyst M.S. in Geographic and Cartographic Sciences Use GIS
More informationARIC Manuscript Proposal # PC Reviewed: _9/_25_/06 Status: A Priority: _2 SC Reviewed: _9/_25_/06 Status: A Priority: _2
ARIC Manuscript Proposal # 1186 PC Reviewed: _9/_25_/06 Status: A Priority: _2 SC Reviewed: _9/_25_/06 Status: A Priority: _2 1.a. Full Title: Comparing Methods of Incorporating Spatial Correlation in
More informationTypes of spatial data. The Nature of Geographic Data. Types of spatial data. Spatial Autocorrelation. Continuous spatial data: geostatistics
The Nature of Geographic Data Types of spatial data Continuous spatial data: geostatistics Samples may be taken at intervals, but the spatial process is continuous e.g. soil quality Discrete data Irregular:
More informationCombining Incompatible Spatial Data
Combining Incompatible Spatial Data Carol A. Gotway Crawford Office of Workforce and Career Development Centers for Disease Control and Prevention Invited for Quantitative Methods in Defense and National
More informationGIS in Locating and Explaining Conflict Hotspots in Nepal
GIS in Locating and Explaining Conflict Hotspots in Nepal Lila Kumar Khatiwada Notre Dame Initiative for Global Development 1 Outline Brief background Use of GIS in conflict study Data source Findings
More informationSpatial correlation and demography.
Spatial correlation and demography. Sébastien Oliveau, Christophe Guilmoto To cite this version: Sébastien Oliveau, Christophe Guilmoto. Spatial correlation and demography.: Exploring India s demographic
More informationExploratory Spatial Data Analysis (And Navigating GeoDa)
Exploratory Spatial Data Analysis (And Navigating GeoDa) June 9, 2006 Stephen A. Matthews Associate Professor of Sociology & Anthropology, Geography and Demography Director of the Geographic Information
More informationA Micro-Spatial Analysis of Violent Crime
A Micro-Spatial Analysis of Violent Crime The Experience of a Small Southern City Isaac T. Van Patten Radford University Levels of Measurement Macro-spatial analysis Crime at the state, regional, or national
More informationGIST 4302/5302: Spatial Analysis and Modeling
GIST 4302/5302: Spatial Analysis and Modeling Basics of Statistics Guofeng Cao www.myweb.ttu.edu/gucao Department of Geosciences Texas Tech University guofeng.cao@ttu.edu Spring 2015 Outline of This Week
More informationApplication of eigenvector-based spatial filtering approach to. a multinomial logit model for land use data
Presented at the Seventh World Conference of the Spatial Econometrics Association, the Key Bridge Marriott Hotel, Washington, D.C., USA, July 10 12, 2013. Application of eigenvector-based spatial filtering
More informationOverview of Statistical Analysis of Spatial Data
Overview of Statistical Analysis of Spatial Data Geog 2C Introduction to Spatial Data Analysis Phaedon C. Kyriakidis www.geog.ucsb.edu/ phaedon Department of Geography University of California Santa Barbara
More informationWhat s special about spatial data?
What s special about spatial data? Road map Geographic Information analysis The need to develop spatial thinking Some fundamental geographic concepts (PBCS) What are the effects of space? Spatial autocorrelation
More informationUsing GIS to Identify Pedestrian- Vehicle Crash Hot Spots and Unsafe Bus Stops
Using GIS to Identify Pedestrian-Vehicle Crash Hot Spots and Unsafe Bus Stops Using GIS to Identify Pedestrian- Vehicle Crash Hot Spots and Unsafe Bus Stops Long Tien Truong and Sekhar V. C. Somenahalli
More informationMeasuring Geographic Access to Primary Care Physicians
Measuring Geographic Access to Primary Care Physicians The New Mexico Health Policy Commission and the University of New Mexico s Division of Government Research have been working cooperatively to collect
More informationObjectives Define spatial statistics Introduce you to some of the core spatial statistics tools available in ArcGIS 9.3 Present a variety of example a
Introduction to Spatial Statistics Opportunities for Education Lauren M. Scott, PhD Mark V. Janikas, PhD Lauren Rosenshein Jorge Ruiz-Valdepeña 1 Objectives Define spatial statistics Introduce you to some
More informationAnalyzing the Geospatial Rates of the Primary Care Physician Labor Supply in the Contiguous United States
Analyzing the Geospatial Rates of the Primary Care Physician Labor Supply in the Contiguous United States By Russ Frith Advisor: Dr. Raid Amin University of W. Florida Capstone Project in Statistics April,
More informationIdentifying Megaregions in the US: Implications for Infrastructure Investment
7. 10. 2 0 08 Identifying Megaregions in the US: Implications for Infrastructure Investment Dr. Myungje Woo Dr. Catherine L. Ross Jason Barringer Harry West Jessica Lynn Harbour Doyle Center for Quality
More informationBook Review: A Social Atlas of Europe
Book Review: A Social Atlas of Europe Ferreira, J Author post-print (accepted) deposited by Coventry University s Repository Original citation & hyperlink: Ferreira, J 2015, 'Book Review: A Social Atlas
More informationMedical GIS: New Uses of Mapping Technology in Public Health. Peter Hayward, PhD Department of Geography SUNY College at Oneonta
Medical GIS: New Uses of Mapping Technology in Public Health Peter Hayward, PhD Department of Geography SUNY College at Oneonta Invited research seminar presentation at Bassett Healthcare. Cooperstown,
More informationJohns Hopkins University Fall APPLIED ECONOMICS Regional Economics
Johns Hopkins University Fall 2017 Applied Economics Sally Kwak APPLIED ECONOMICS 440.666 Regional Economics In this course, we will develop a coherent framework of theories and models in the field of
More informationThe polygon overlay problem in electoral geography
The polygon overlay problem in electoral geography Romain Louvet *1,2, Jagannath Aryal 2, Didier Josselin 1,3, Christèle Marchand-Lagier 4, Cyrille Genre-Grandpierre 1 1 UMR ESPACE 7300 CNRS, Université
More informationGlobal activity distribution patterns of top international Chinese contractors Chuan Chen1, a, Hongjiang Li1, b and Igor Martek2, c
International Conference on Management Science and Innovative Education (MSIE 2015) Global activity distribution patterns of top international Chinese contractors Chuan Chen1, a, Hongjiang Li1, b and Igor
More informationOutline. Practical Point Pattern Analysis. David Harvey s Critiques. Peter Gould s Critiques. Global vs. Local. Problems of PPA in Real World
Outline Practical Point Pattern Analysis Critiques of Spatial Statistical Methods Point pattern analysis versus cluster detection Cluster detection techniques Extensions to point pattern measures Multiple
More informationLecture 9: Geocoding & Network Analysis
Massachusetts Institute of Technology - Department of Urban Studies and Planning 11.520: A Workshop on Geographic Information Systems 11.188: Urban Planning and Social Science Laboratory Lecture 9: Geocoding
More informationUnderstanding the modifiable areal unit problem
Understanding the modifiable areal unit problem Robin Flowerdew School of Geography and Geosciences, University of St Andrews March 2009 Acknowledgements Mick Green (Lancaster) and David Steel (Wollongong),
More informationTime: the late arrival at the Geocomputation party and the need for considered approaches to spatio- temporal analyses
Time: the late arrival at the Geocomputation party and the need for considered approaches to spatio- temporal analyses Alexis Comber 1, Paul Harris* 2, Narumasa Tsutsumida 3 1 School of Geography, University
More informationSpatio-temporal Small Area Analysis for Improved Population Estimation Based on Advanced Dasymetric Refinement
Spatio-temporal Small Area Analysis for Improved Population Estimation Based on Advanced Dasymetric Refinement Hamidreza Zoraghein, Stefan Leyk, Barbara Buttenfield and Matt Ruther ABSTRACT: Demographic
More informationComputing error measures for migration distance estimates in historical linked data sets
Computing error measures for migration distance estimates in historical linked data sets Rebecca Vick, Minnesota Population Center, University of Minnesota Sula Sarkar, Minnesota Population Center, University
More information2/7/2018. Module 4. Spatial Statistics. Point Patterns: Nearest Neighbor. Spatial Statistics. Point Patterns: Nearest Neighbor
Spatial Statistics Module 4 Geographers are very interested in studying, understanding, and quantifying the patterns we can see on maps Q: What kinds of map patterns can you think of? There are so many
More informationThe Building Blocks of the City: Points, Lines and Polygons
The Building Blocks of the City: Points, Lines and Polygons Andrew Crooks Centre For Advanced Spatial Analysis andrew.crooks@ucl.ac.uk www.gisagents.blogspot.com Introduction Why use ABM for Residential
More informationBasics of Geographic Analysis in R
Basics of Geographic Analysis in R Spatial Autocorrelation and Spatial Weights Yuri M. Zhukov GOV 2525: Political Geography February 25, 2013 Outline 1. Introduction 2. Spatial Data and Basic Visualization
More informationLinking Industry and Occupation Clusters in Regional Economic Development
Linking Industry and Occupation Clusters in Regional Economic Development Charting the Course for Regional Development: First Annual EDA Economic Development Research Symposium Clarion Hotel Morgan Morgantown,
More informationR E SEARCH HIGHLIGHTS
Canada Research Chair in Urban Change and Adaptation R E SEARCH HIGHLIGHTS Research Highlight No.8 November 2006 THE IMPACT OF ECONOMIC RESTRUCTURING ON INNER CITY WINNIPEG Introduction This research highlight
More informationField Course Descriptions
Field Course Descriptions Ph.D. Field Requirements 12 credit hours with 6 credit hours in each of two fields selected from the following fields. Each class can count towards only one field. Course descriptions
More informationIn matrix algebra notation, a linear model is written as
DM3 Calculation of health disparity Indices Using Data Mining and the SAS Bridge to ESRI Mussie Tesfamicael, University of Louisville, Louisville, KY Abstract Socioeconomic indices are strongly believed
More informationPolicy Paper Alabama Primary Care Service Areas
Aim and Purpose Policy Paper Alabama Primary Care Service Areas Produced by the Office for Family Health Education & Research, UAB School of Medicine To create primary care rational service areas (PCSA)
More informationA spatial literacy initiative for undergraduate education at UCSB
A spatial literacy initiative for undergraduate education at UCSB Mike Goodchild & Don Janelle Department of Geography / spatial@ucsb University of California, Santa Barbara ThinkSpatial Brown bag forum
More informationNational Statistics 2001 Area Classifications
National Statistics 2001 Area Classifications John Charlton, ONS see http://neighbourhood.statistics.gov.uk areaclassifications@ons.gov.uk Copyright ONS What are the Area Classifications Summarise 2001
More informationA User s Guide to the Federal Statistical Research Data Centers
A User s Guide to the Federal Statistical Research Data Centers Mark Roberts Professor of Economics and Director PSU FSRDC September 2016 M. Roberts () RDC User s Guide September 2016 1 / 14 Outline Introduction
More information