Spatial statistics is a rapidly growing field due to the increased availability of large amounts of data within areas such as global climate monitoring, disease surveillance, and image data collection. Balancing theoretical and methodological techniques for the analysis of spatial statistics, this book fills an existing gap in the spatial statistics literature by presenting a unique approach to modeling and analyzing high-dimensional and spatially dependent data. This book features theoretical concepts alongside the needed methodology in order to provide readers with a better understanding of the computational techniques used to model large spatial data sets. Providing a foundation of spatial data analysis from a machine learning perspective supported by statistical inference, this book details approaches to large-to-massive data including variable/feature selection; predictive spatial modeling; spatial functional data analysis; spatial clustering; covariance estimation; and the application of kriging. The authors have a primary focus on non-Bayesian theory and methods, but do also address Bayesian methodology for analyzing spatial data as needed. Topical coverage includes: spatial data; basic kriging; covariance models; asymptotic theory for spatial statistics; variable selection and clustering for independents data; covariance estimation for large spatial data; likelihood-based methods; tapered covariance estimation; blocking; low-rank estimation; linear and additive model selection; generalized linear models; inferential issues after model selection; predictive model building for large spatial data; simultaneous estimation of mean and variance; dynamic spatial data; spatial functional data; spatial data on irregular surface; spatial data from image analysis; and spatial clustering.