Spatial Analysis Using Big Data

5.7. Methods for large data

5.7.1. Outline

In recent years, spatially indexed large or even massive data are becoming available. Researchers try to refine spatial econometrics models to be applicable to such data. The following subsections briefly overview such an attempt.

Before that, we would like to mention Arbia et al. (2019b), who reported the results of a systematic simulation study on the limits of the current methodologies when estimating spatial models with large datasets. They simulate an SLM, and estimated it using two-stage least squares (2SLS) (see the next subsection), ML, and Bayesian estimator. They found that for the last two methods, it is not possible to estimate models with more than 10,000 units when matrices are characterized ...