ANALYSIS OF QUALITATIVE DATA USING GLIM: TWO EXAMPLES BASED ON SHOPPING SURVEY DATA

We recommend use of the widely available computer package GLIM to analyze relationships involving categorical, or qualitative, variables. The package is flexible and fairly simple to use and readily produces data plots as well as information on parameter estimates and residual values. In order to explain how to fit log-linear and logistic-linear models using GLIM, we give two examples based on sample survey data from grocery shoppers in Oxford, England. In the first, we examine relationships between a categorical response to an attitude statement and categorical explanatory variables using the log-linear model and, in the second, relationships between use and a mixture of continuous and categorical explanatory variables using the logistic-linear (logit) model.     

ANALYSIS OF COUNT DATA USING POISSON REGRESSION*

In geographical research the data of interest are often in the form of counts. Standard regression analysis is inappropriate for such data, but if certain assumptions are met, a form of regression based on the Poisson distribution can be used. This paper illustrates the use of Poisson regression in the computer package GLIM with an example from historical geography. Apprentice migration to Edinburgh is regressed on a combination of categorical, count, and continuous explanatory variables.     

Spatial crime distribution and prediction for sporting events using social media

ABSTRACT

Sporting events attract high volumes of people, which in turn leads to increased use of social media. In addition, research shows that sporting events may trigger violent behavior that can lead to crime. This study analyses the spatial relationships between crime occurrences, demographic, socio-economic and environmental variables, together with geo-located Twitter messages and their ‘violent’ subsets. The analysis compares basketball and hockey game days and non-game days. Moreover, this research aims to analyze crime prediction models using historical crime data as a basis and then introducing tweets and additional variables in their role as covariates of crime. First, this study investigates the spatial distribution of and correlation between crime and tweets during the same temporal periods. Feature selection models are applied in order to identify the best explanatory variables. Then, we apply localized kernel density estimation model for crime prediction during basketball and hockey games, and on non-game days. Findings from this study show that Twitter data, and a subset of violent tweets, are useful in building prediction models for the seven investigated crime types for home and away sporting events, and non-game days, with different levels of improvement.     

基于Weibull函数和Gamma函数的环境污染与经济增长的关系

环境库兹涅茨曲线是研究环境污染与经济增长之间关系的有效工具。传统的环境库兹涅茨曲线研究大多采用线性模型或对数线性模型,其在采用模型描述关系的问题上存在一定的不足。为克服这些不足,本文采用Weibull函数和Gamma函数形式的面板数据模型对中国29个省区1989~2005年四种环境污染指标人均排放量与人均收入之间的关系予以研究。结果表明Weibull函数和Gamma函数的面板数据模型拟合效果较好,且参数具有较好的解释能力;人均废水和人均SO₂都随人均收入增加先上升后减少,在25000元附近出现结构转变点,而人均固体废弃物和人均废气随人均收入增加则呈现单调增加的变化趋势,没有出现结构转变点。     

Generalizing Poisson Regression: Including Apriori Information Using the Method of Offsets*

In a recent note in The Professional Geographer, Lovett and Flowerdew (1989) introduce the analysis of count data using Poisson regression. Concern here rests with generalizing their discussion to show how a priori information may be included into models that incorporate dependent variables enumerated as counts. These models can be calibrated using the OFFSET command in GLIM. The paper concludes with a brief discussion of the applicability of the method of offsets to other problems in spatial analysis.     

Modelling landslide distribution on loess soils in China: an investigation

Abstract

Catastrophic landslides are a severe environmental problem on the Loess Plateau of China. In order to establish which slopes are most at risk, their spatial incidence can be modelled using geographical information systems techniques together with a database describing the past landslides. Three modelling strategies, sieve mapping, weighting factors and log-linear modelling, are adopted and compared. It is suggested that the log-linear modelling approach is the most general and makes best use of the available information.     

Mapping illegal domestic waste disposal potential to support waste management efforts in Queensland,Australia

Illegal disposal of waste is a significant management issue for contemporary governments because of the hazards posed to both human and ecosystem health. Understanding the complex distribution pattern of illegal waste and the range of economic, environmental and social factors influencing this distribution is valuable for improving the effectiveness and efficiency of waste management efforts. This article examines the applicability of mapping illegal waste disposal in the Sunshine Coast (Queensland, Australia) through the identification and integration of predictive spatial data in a geographic information system. A statistical model of illegal waste disposal was developed using a binary logistic regression analysis to identify explanatory variables suitable for predicting the distribution of illegal waste. Five statistically significant explanatory variables were identified through this analysis: population density, primary land use, distance to the nearest road, waste facility and roadside amenity. The generated statistical model had a predictive success of 86.1% with all indicators suggesting good model fit (χ² = 474.3, P = 0 with df = 22) across the study area. Standardised spatial data on each explanatory variable were combined using a weighted linear combination analysis and the results were classified into five categories from very low to very high illegal waste disposal potentials using the equal interval method. The resultant mapping identified 6.9% of the study area as having very high illegal waste disposal potential, and subsequent validation indicated that 32.9% of known illegal waste disposal sites were located within these areas.     

Extended weights of evidence modeling for the pseudo-estimation of metal grades

The weights of evidence modeling (WEM) for binary patterns is extended to take account of general categorical variables. The extension makes it possible to use the weights of evidence model in estimating the conditional probability distributions of metal grades. First, the target feature is converted into a set of binary target indicators. Second, the posterior probabilities are estimated for each of the target categories. Third, the estimates are combined to yield the posterior probability distribution of the target feature. Finally, the pseudometal estimates are derived from the probability distribution. The metal grade estimates are prefixed with pseudo, because the estimates are created from indirect evidence (explanatory variables). The pseudo-estimates provide a unique quantitative means to the delineation of exploration targets. This advantage reduces the ambiguities of target selection based solely on probability estimates. In order to use the generalized WEM, continuous geoscience attributes must be converted into categorical variables by means of optimal segmentation based on the target attribute of interest. The segmentation may be viewed as a process of defining evidence of the target feature. The extended weights of evidence model is demonstrated on a case study to select gold targets of Carlin type. The dataset used in the modeling includes apparent resistivity fields, soil geochemical samples, lithological and alteration information, and structural data.     

The Origin and Pattern Of Development Of Urban Black Social Areas

Abstract

This exercise has been used successfully by more than a thousand students is an introductory college course, and it probably could be equally successful with high school students. Census tract data are used in a simulated research project. Quality of housing in the dependent variable (whose distribution we wish to “explain”), with owner-occupancy and age of housing as independent (or “explanatory”) variables. The distribution of the three variables is compared visually (by maps) and statistically (by scattergrams). A regression line is placed on the better scattergram, and residual values are mapped to provide a basis for the formulation of better working hypotheses.     

Development and test of an error model for categorical data

Abstract

An error model for spatial databases is defined here as a stochastic process capable of generating a population of distorted versions of the same pattern of geographical variation. The differences between members of the population represent the uncertainties present in raw or interpreted data, or introduced during processing. Defined in this way, an error model can provide estimates of the uncertainty associated with the products of processing in geographical information systems. A new error model is defined in this paper for categorical data. Its application to soil and land cover maps is discussed in two examples: the measurement of area and the measurement of overlay. Specific details of implementation and use are reviewed. The model provides a powerful basis for visualizing error in area class maps, and for measuring the effects of its propagation through processes of geographical information systems.     

Computational improvements to multi-scale geographically weighted regression

ABSTRACT

Geographically Weighted Regression (GWR) has been broadly used in various fields to model spatially non-stationary relationships. Multi-scale Geographically Weighted Regression (MGWR) is a recent advancement to the classic GWR model. MGWR is superior in capturing multi-scale processes over the traditional single-scale GWR model by using different bandwidths for each covariate. However, the multiscale property of MGWR brings additional computation costs. The calibration process of MGWR involves iterative back-fitting under the additive model (AM) framework. Currently, MGWR can only be applied on small datasets within a tolerable time and is prohibitively time-consuming to run with moderately large datasets (greater than 5,000 observations). In this paper, we propose a parallel implementation that has crucial computational improvements to the MGWR calibration. This improved computational method reduces both memory footprint and runtime to allow MGWR modelling to be applied to moderate-to-large datasets (up to 100,000 observations). These improvements are integrated into the mgwr python package and the MGWR 2.0 software, both of which are freely available to download.     

Multi-label class assignment in land-use modelling

During the last two decades, a variety of models have been applied to understand and predict changes in land use. These models assign a single-attribute label to each spatial unit at any particular time of the simulation. This is not realistic because mixed use of land is quite common. A more detailed classification allowing the modelling of mixed land use would be desirable for better understanding and interpreting the evolution of the use of land. A possible solution is the multi-label (ML) concept where each spatial unit can belong to multiple classes simultaneously. For example, a cluster of summer houses at a lake in a forested area should be classified as water, forest and residential (built-up). The ML concept was introduced recently, and it belongs to the machine learning field. In this article, the ML concept is introduced and applied in land-use modelling. As a novelty, we present a land-use change model that allows ML class assignment using the k nearest neighbour (kNN) method that derives a functional relationship between land use and a set of explanatory variables. A case study with a rich data-set from Luxembourg using biophysical data from aerial photography is described. The model achieves promising results based on the well-known ML evaluation criteria. The application described in this article highlights the value of the multi-label k nearest neighbour method (MLkNN) for land-use modelling.     

Sources of Uncertainty in Shear Stress and Roughness Length Estimates Derived from Velocity Profiles∗

Shear stress is a fundamental parameter in many sediment transport expressions. It is commonly estimated from information contained in the velocity profile, specifically, shear velocity, u _?, and roughness length, z ₀ Under ideal conditions, the functional relationship between elevation above the bed and flow velocity is log-linear, as expressed by the “law of the wall.” Unless the field data conform exactly to this ideal relationship, there is uncertainty in estimates of u _? and z ₀. derived from velocity profiles. This uncertainty depends on the degree of scatter or deviation from the assumed log-linear model. Expressions are presented to quantify the error and to correct for it. Guidelines are suggested to minimize potential uncertainty, especially with regard to instrument deployment and velocity profile analysis. Calculation of confidence intervals around estimates of u _? and z ₀, as proposed by Wilkinson (1984), is necessary and recommended.     

Using Dummy Variables to Estimate Economic Base Multipliers

Economic base analysis is a cost-effective and accurate means of predicting employment impacts in relatively small and uncomplicated regional (community) economies. Extending previous research using the Arizona Community Data Set, this paper estimates economic base multipliers in different types of communities by introducing dummy variables into regression equations. Total employment, transfer payments, and population potential are used as explanatory variables. Manufacturing centers are shown to have larger multipliers than diversified places, mining settlements, service and trade communities, and utility towns.     

Comparison of landslide susceptibility based on a decision-tree model and actual landslide occurrence: The Akaishi Mountains, Japan

This paper proposes a statistical decision-tree model to analyze landslide susceptibility in a wide area of the Akaishi Mountains, Japan. The objectives of this study were to validate the decision-tree model by comparing landslide susceptibility and actual landslide occurrence, and to reveal the relationships among landslide occurrence, topography, and geology. Landslide susceptibility was examined through ensemble learning with a decision tree. Decision trees are advantageous in that estimation processes and order of important explanatory variables are explicitly represented by the tree structures. Topographic characteristics (elevation, slope angle, profile curvature, plan curvature, and dissection and undissection height) and geological data were used as the explanatory variables. These topographic characteristics were calculated from digital elevation models (DEMs). The objective variables were landslide occurrence and reactivation data between 1992 and 2002 that were depicted by satellite image analysis. Landslide susceptibility was validated by comparing actual data on landslides that occurred and reactivated after the model was constructed (between 2002 and 2004).This study revealed that, from 2002 to 2004, landslides tended to occur and reactivate in catchments with high landslide susceptibility. The landslide susceptibility map thus depicts the actual landslide occurrence and reactivation in the Akaishi Mountains. This result indicates that the decision-tree model has appropriate accuracy for estimating the probabilities of future landslides. The tree structure indicates that landslides occurred and reactivated frequently in the catchments that had an average slope angle exceeding ca. 29° and a mode of slope angle exceeding 33°, which agree well with previous studies. A decision tree also quantitatively expresses important explanatory variables at the higher order of the tree structure.     

Analyzing Changes in Urban Form and Commuting Time*

Abstract

Urban work trips have changed in important ways during the last decades. In Québec City, a medium-sized Canadian metropolitan area, commuting distances increased for both male and female workers between 1977 and 1996, while durations increased for male workers and decreased for female workers. This article seeks to identify spatial and social factors responsible for these changes. We develop a disaggregate model of trip duration estimated on the basis of large samples derived from travel surveys comparable through time. Using categorical variables to specify change, we are able to separate change effects from level effects attributable to various dimensions of urban form. Our analysis clearly indicates that, once travel mode and key social factors are controlled for, the shift from a monocentric to a dispersed city form is responsible, in the Québec metropolitan area, for increasing commuting time. This is contrary to findings in larger metropolitan areas, where, it has been argued, the suburbanization of jobs maintains stability in commuting duration.     

HyperArc: a task-oriented hypertext GIS interface

Abstract

The rationale, development and -implementation of a task-oriented hypertext GIS interface (HyperArc) are described. HyperArc is a fully functional interface on an Apple Macintosh computer acting as a client to a host computer running the ARC/INFO GIS. The package has been developed using the HyperCard package, and makes full use of the Apple Macintosh WIMP interface, enabling the user to perform a range of standard ARC/INFO tasks, including a variety of map displays and data queries. The key advantage of this kind of interface is that it permits the creation of tasks or sets of spatial procedures which can be easily customised and executed by the end user without any knowledge of the system command language.     

Efficiently implementable algebra for distributed in-network spatial analysis

Existing sensor network query processors (SNQPs) have demonstrated that in-network processing is an effective and efficient means of interacting with wireless sensor networks (WSNs) for data collection tasks. Inspired by these findings, this article investigates the question as to whether spatial analysis over WSNs can be built upon established distributed query processing techniques, but, here, emphasis is on the spatial aspects of sensed data, which are not adequately addressed in the existing SNQPs. By spatial analysis, we mean the ability to detect topological relationships between spatially referenced entities (e.g. whether mist intersects a vineyard or is disjoint from it) and to derive representations grounded on such relationships (e.g. the geometrical extent of that part of a vineyard that is covered by mist). To support the efficient representation, querying and manipulation of spatial data, we use an algebraic approach. We revisit a previously proposed centralized spatial algebra comprising a set of spatial data types and a comprehensive collection of operations. We have redefined and re-conceptualized the algebra for distributed evaluation and shown that it can be efficiently implemented for in-network execution. This article provides rigorous, formal definitions of the spatial data types, points, lines and regions, together with spatial-valued and topological operations over them. The article shows how the algebra can be used to characterize complex and expressive topological relationships between spatial entities and spatial phenomena that, due to their dynamic, evolving nature, cannot be represented a priori.     

Aspects of the implementation of the GEOVIEW design

Abstract

GEOVIEW, an integrated system, uses relational database technology and a graphics package (Graphical Kernel System) to offer a flexible environment in which to develop applications of geographical information systems (GIS). It has facilities to represent data in different spatial data structures. Data are stored and retrieved efficiently by using variable length raw data. The unified representation offers the benefit of storing entities in a single relation and eases the process of overlaying different entities. GEOVIEW also provides a mechanism for tailoring user interfaces to suit the needs of different applications by means of a facility to generate macros and menus.

Processing requirements for GIS applications can be supported by using tools provided by relational database technology and by graphics packages. High-level language interfaces which can process dynamic statements and bind dynamic variables are needed to develop an efficient database interface module. A graphics segment facility is essential to provide editing functions and to maximize the use of the local processing power of graphics workstations in the graphics interface module. Further improvements in performance can be made by using the array fetch facility and linear keys for spatial searching.