| Course Content: Spatial autocorrelation refresher, spatial weights, non-traditional spatial weights, semi-parametric approaches in spatial econometrics, including different ways to introduce semi-parametrics into spatial models, and non-parametric spatial covariance estimation.
Maximum Likelihood (ML) and General Method of Moments (GMM) approaches in spatial econometrics, including Lagrange Multiplier tests, ML estimation, optimal GMM, models for both heteroskedasticity and spatial error dependence and heteroskedastic and autocorrelation consistent (HAC) variance estimators This course deals with estimation methods for spatial econometric models belonging to the general category of Generalized Method of Moments. It starts with a brief refresher of Instrumental Variables, 2SLS and the basic principles of Method of Moments estimation. Next, It proceeds with more specialized materials covering the application of Generalized Methods of Moments to the estimation and testing of spatial models.
Outline of the course:
1. IV and GMM review
1.1. Instrumental Variables Estimation
1.2. Generalized Method of Moments
2. GM and IV/GMM
2.1. Spatially Weighted Least Squares
2.2. GM Estimation for Spatial Error Models
2.3. Heteroskedasticity Robust GMM Estimation
2.4. IV and GMM Spatial Lag Models
This workshop will introduce techniques for spatial vector data analysis (areas, points, lines) in ecology and conservation science. Mapping, measuring, and spatially exploring and analysing vector data is an essential component of research in the fields of landscape ecology, biodiversity conservation, forest ecology, and wildlife biology including the study of animal movement and human-wildlife conflicts in modern landscapes. The workshop aims to develop the ability to gather, use, and analyse spatial vector data among students and scientists in the field of ecology and conservation in India. This course provides a survey of techniques of spatial data analysis, covering geovisualization, point pattern analysis, variogram analysis, kriging and spatial autocorrelation. The emphasis is on gaining a solid understanding of the merits of each of the methods and an appreciation of when they should be applied. The level of mathematical treatment will be intermediate, with some mathematical expressions, but no theorems and proofs. The main goal of the class is for you to become familiar with the essential methodological and practical issues that are involved in carrying out applied spatial data analysis. This course provides a survey of techniques of spatial data analysis, covering geovisualization, point pattern analysis, variogram analysis, kriging and spatial autocorrelation. The emphasis is on gaining a solid understanding of the merits of each of the methods and an appreciation of when they should be applied. The level of mathematical treatment will be intermediate, with some mathematical expressions, but no theorems and proofs. The main goal of the class is for you to become familiar with the essential methodological and practical issues that are involved in carrying out applied spatial data analysis.
(No news has been posted yet)
| This is the moodle development site of the GeoDa Center. It contains
e-Learning resources focused on spatial data analysis and spatial regression. Special emphasis is on supporting advanced methods and software tools for spatial econometric modeling of panel data and binary response models (spatial probit). Methods covered include instrumental variables estimation, General Method of Moments, non-parametric and semi-parametric techniques as well as Bayesian approaches. |
