Loved by learners at thousands of companies
Course Description
Everything happens somewhere, and increasingly the place where all these things happen is being recorded in a database. There is some truth behind the oft-repeated statement that 80% of data have a spatial component. So what can we do with this spatial data? Spatial statistics, of course! Location is an important explanatory variable in so many things - be it a disease outbreak, an animal's choice of habitat, a traffic collision, or a vein of gold in the mountains - that we would be wise to include it whenever possible. This course will start you on your journey of spatial data analysis. You'll learn what classes of statistical problems present themselves with spatial data, and the basic techniques of how to deal with them. You'll see how to look at a mess of dots on a map and bring out meaningful insights.
For Business
Training 2 or more people?
Get your team access to the full DataCamp library, with centralized reporting, assignments, projects and more- 1
Introduction
FreeAfter a quick review of spatial statistics as a whole, you'll go through some point-pattern analysis. You'll learn how to recognize and test different types of spatial patterns.
Problems in spatial statistics50 xpSimple spatial principles100 xpPlotting areas100 xpUniform in a circle100 xpSimulation and testing with spatstat50 xpQuadrat count test for uniformity100 xpCreating a uniform point pattern with spatstat100 xpSimulating clustered and inhibitory patterns100 xpPoint pattern testing50 xpFurther testing50 xpNearest-neighbor distributions100 xpOther point pattern distribution functions100 xpTree location pattern50 xp - 2
Point Pattern Analysis
Point Pattern Analysis answers questions about why things appear where they do. The things could be trees, disease cases, crimes, lightning strikes - anything with a point location.
Bivariate point patterns50 xpCrime in Preston100 xpViolent crime proportion estimation100 xpSpatial segregation50 xpBandwidth selection100 xpSegregation probabilities100 xpMapping segregation100 xpSpace-time data50 xpSasquatch data100 xpSpatial pattern of bigfoot sightings50 xpTemporal pattern of bigfoot sightings100 xpTemporal pattern analysis50 xpSpace-time clustering50 xpPreparing data for space-time clustering100 xpMonte-carlo test of space-time clustering100 xpSpace-time clustering explanation50 xp - 3
Areal Statistics
So much data is collected in administrative divisions that there are specialized techniques for analyzing them. This chapter presents several methods for exploring data in areas.
Areal statistics50 xpLondon EU referendum data100 xpCartogram100 xpSpatial autocorrelation test100 xpSpatial health data50 xpLondon health data100 xpBinomial confidence intervals100 xpExceedance probabilities100 xpGeneralized linear models in space50 xpA Poisson GLM100 xpResiduals100 xpCorrelation in spatial GLMs50 xpFit a Bayesian GLM100 xpAdding a spatially autocorrelated effect100 xpMapping the spatial effects100 xp - 4
Geostatistics
Originally developed for the mining industry, geostatistics covers the analysis of location-based measurement data. It enables model-based interpolation of measurements with uncertainty estimation.
Geostatistical data50 xpCanadian geochemical survey data100 xpFitting a trend surface100 xpPredicting from a trend surface100 xpThe variogram50 xpVariogram estimation100 xpVariogram with spatial trend100 xpVariogram model fitting100 xpKriging predictions50 xpFilling in the gaps100 xpMaking a prediction grid100 xpGridded predictions100 xpAutomatic kriging50 xpAuto-kriging at point locations100 xpAuto-kriging over a grid100 xpLast words50 xp
For Business
Training 2 or more people?
Get your team access to the full DataCamp library, with centralized reporting, assignments, projects and moreDatasets
Canadian geological survey soil acidityBounding region for ca_geo.rdsFlu incidents by London borough, 2017EU referendum voting by London borough, 2016An OpenStreetMap map of Preston, UKCrime in Preston, UKPh gridSasquatch sightings, 1990 to 2016Collaborators
Barry Rowlingson
See More
Research Fellow at Lancaster University
Barry Rowlingson is a Research Fellow in the Lancaster Medical School, part of the Faculty of Health and Medicine at Lancaster University. His primary field of research is statistical applications in epidemiology and other health-related areas. He plays assorted instruments, takes pictures in exotic places, and drives an old Land Rover.
Join over 13 million learners and start Spatial Statistics in R today!
Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.