Spatial econometrics in an age of CyberGIScience

Abstract

In this article, we focus on the evolution of the technology that lies at the basis of implementing spatial econometric methods into software tools. We review the changing methodological emphases and their implications for data structures and computational infrastructure required for estimation and inference. We review the evolution of software solutions, starting with SpaceStat and GeoDa and moving on to the PySAL open source library of spatial analytical functions (Rey and Anselin 2010, PySAL: a Python library of spatial analytical methods. In: M.M. Fischer and A. Getis, eds. Handbook of applied spatial analysis. Berlin: Springer,[175][176][177][178][179][180][181][182][183][184][185][186][187][188][189][190][191][192][193]. We compare these approaches with other software solutions, such as the R spatial analytical routines and recently released Stata functionality for spatial econometrics. We follow the review with a discussion of requirements and challenges encountered when moving these software tools into a CyberGIScience framework. We focus on the efficient data structures, the need for metadata and provenance tracking, as well as high-performance computing requirements. We close with the outline of a vision for a 'spatial econometrics workbench' as a core component of cyberinfrastructure for GIScience.

Open Geospatial Analytics with PySAL

“…This is not a simple matter of installing the desktop version of the PySAL library on new hardware, but rather a number of fundamental challenges arise that must be addressed if the potential of high performance computing (HPC) for spatial analysis are to be realized. Key among these are the issues of provenance and meta-data for spatial analytical workflows, which become particularly critical when considering distributed computing and interoperability between different services [16]. While there has been much work on provenance and meta-data for spatial data and geoprocessing, similar frameworks for chaining together spatial analytical methods are in their infancy.…”

“…These primarily concern the ability of the software to scale beyond the case study of the prototype as well as to support interoperability with other spatial analytical services and software, and scientific replication and reproducibility. The scalability challenge arises because most of the spatial statistical and econometric software that is leveraged in these prototype systems was not originally designed for the big data era or to exploit new types of high performance computing environments [16]. Moreover, while calls for replication and reproducibility have recently moved to the forefront of the open science agenda, these concepts were not on the white boards of the designers of our previous generation spatial analytical software.…”

Integrating Spatial Data Linkage and Analysis Services in a Geoportal for China Urban Research

“…Broadly speaking, the attribute and parameter semantics can be incorporated into the data provenance (Yue et al. ) and model provenance (Anselin and Rey, ) architecture. Our work incorporates parameter validation and its meta‐information into the OGC's WPS standard, and supports data‐dependent validation in computation components.…”

Parallel optimal choropleth map classification in PySAL

“…Recent years have seen increased attention directed toward the use of high-performance computing (HPC) in spatial analysis (Armstrong 2000;Openshaw and Turton 2000;Wang 2008; Anselin and Rey 2012). Motivated by the tremendous opportunities that advances in computing technologies hold, this focus seeks to move spatial analysis into a new era of CyberGIScience.…”