# Spatial Microeconometrics

Foundations of spatial microeconometrics modelingA micro-level approach to spatial econometricsAdvantages of spatial microeconometric analysisSources of spatial micro-dataSources of uncertainty in spatial micro-dataConclusions and plan of the bookII: Modeling the spatial behavior of economic agents in a given set of locationsPreliminary definitions and conceptsNeighborhood and the W matrixMoran’s I and other spatial correlation measuresThe Moran scatterplot and local indicators of spatial correlationConclusionsBasic cross-sectional spatial linear modelsIntroductionRegression models with spatial autoregressive componentsPure spatial autoregressionThe spatial error modelThe spatial lag modelThe spatial Durbin modelThe general spatial autoregressive model with spatial autoregressive error structureTest of residual spatial autocorrelation with explicit alternative hypothesesMarginal impactsEffects of spatial imperfections of micro-dataIntroductionMeasurement error in spatial error modelsMeasurement error in spatial lag modelsProblems in regressions on a spatial distanceNon-linear spatial modelsNon-linear spatial regressionsStandard non-linear modelsLogit and probit modelsThe tobit modelSpatial probit and logit modelsModel specificationEstimationThe spatial tobit modelEstimationFurther non-linear spatial modelsMarginal impacts in spatial non-linear modelsSpace–time modelsGeneralitiesFixed and random effects modelsRandom effects spatial modelsFixed effect spatial modelsEstimationIntroductionMaximum likelihoodLikelihood procedures for random effect modelsLikelihood procedures for fixed effect modelsThe generalized method of moments approachGeneralized method of moments procedures for random effects modelsGeneralized method of moments procedures for fixed effects modelsA glance at further approaches in spatial panel data modelingIII: Modeling the spatial locational choices of economic agentsPreliminary definitions and concepts in point pattern analysisSpatial point patterns of economic agentsThe hypothesis of complete spatial randomnessSpatial point processesAggregated point processesInhomogeneous Poisson point processesPoisson cluster point processesRegular point processesClassic exploratory tools and summary statistics for spatial point patternsQuadrat-based methodsDistance-based methodsModels of the spatial location of individualsRipley’s K-functionEstimation of Ripley’s K-functionIdentification of spatial location patternsThe CSR testParameter estimation of the Thomas cluster processParameter estimation of the Matérn cluster processParameter estimation of the log-Gaussian Cox processPoints in a heterogeneous spaceDiggle and Chetwynd’s D-functionBaddeley, Møller and Waagepetersen’s K[sub(inhom)] -functionEstimation of K[sub(inhom)]-functionInference for K[sub(inhom)] -functionMeasuring spatial concentration of industries: Duranton–Overman K-density and Marcon–Puech M-functionDuranton and Overman’s K-densityMarcon and Puech’s M-functionSpace–time modelsDiggle, Chetwynd, Häggkvist and Morris’ space–time K-functionEstimation of space–time K-functionDetecting space–time clustering of economic eventsGabriel and Diggle’s STIK-functionEstimation of STIK-function and inferenceIV: Looking ahead: modeling both the spatial location choices and the spatial behavior of economic agentsFirm demography and survival analysisIntroductionA spatial microeconometric model for firm demographyA spatial model for firm demographyIntroductionThe birth modelThe growth modelThe survival modelA case studyData descriptionThe birth modelThe growth modelThe survival modelConclusionsA spatial microeconometric model for firm survivalIntroductionBasic survival analysis techniquesCase study: The survival of pharmaceutical and medical device manufacturing start-up firms in ItalyData descriptionDefinition of the spatial microeconometric covariatesDefinition of the control variablesEmpirical resultsConclusionAppendicesAppendix 1: Some publicly available spatial datasetsAppendix 2: Creation of a W matrix and preliminary computationsAppendix 4: Non-linear spatial modelsAppendix 5: Space–time modelsAppendix 6: Preliminary definitions and concepts in point pattern analysisAppendix 6.1: Point pattern datasetsAppendix 6.2: Simulating point patternsAppendix 6.2.1: Homogeneous Poisson processesAppendix 6.2.2: Inhomogeneous Poisson processesAppendix 6.2.3: Cox processesAppendix 6.2.4: Poisson cluster processesAppendix 6.2.5: Regular processesAppendix 6.3: Quadrat-based analysisAppendix 6.4: Clark–Evans testAppendix 7: Models of the spatial location of individualsAppendix 7.1: K-function-based CSR testAppendix 7.2: Point process parameters estimation by the method of minimum contrastAppendix 8: Points in a heterogeneous spaceAppendix 8.1: D-function-based test of spatial interactionsAppendix 8.3: Duranton–Overman K-density and Marcon–Puech M-functionAppendix 9: Space–time modelsAppendix 9.1: Space–time K-functionAppendix 9.2: Gabriel and Diggle’s STIK-functionBibliography