Spatial predictive modeling (SPM) is an emerging discipline in applied
sciences, playing a key role in the generation of spatial predictions in
various disciplines. SPM refers to preparing relevant data, developing
optimal predictive models based on point data, and then generating
spatial predictions. This book aims to systematically introduce the
entire process of SPM as a discipline. The process contains data
acquisition, spatial predictive methods and variable selection,
parameter optimization, accuracy assessment, and the generation and
visualization of spatial predictions, where spatial predictive methods
are from geostatistics, modern statistics, and machine learning.
The key features of this book are:
-Systematically introducing major components of SPM process.
-Novel hybrid methods (228 hybrids plus numerous variants) of modern
statistical methods or machine learning methods with mathematical and/or
univariate geostatistical methods.
-Novel predictive accuracy-based variable selection techniques for
spatial predictive methods.
-Predictive accuracy-based parameter/model optimization.
-Reproducible examples for SPM of various data types in R.
This book provides guidelines, recommendations, and reproducible
examples for developing optimal predictive models by considering various
components and associated factors for quality-improved spatial
predictions. It provides valuable tools for researchers, modelers, and
university students not only in SPM field but also in other predictive
modeling fields.
Dr Li has produced over 100 various publications in spatial predictive
modelling, statistical computing, ecological and environmental
modelling, and ecology, developed a number of hybrid methods for SPM,
and published four R packages for variable selections as well as SPM.