1.3 Spatial data

Three main categories are usually used to distinguish between types of spatial data:

Point process data:

consist of locations of points $x_i$ on a given domain, e.g. trees in a forest, cancer cases in Lancashire. The task is usually to describe the distribution and relationship between the $x_i$.

Geostatistical data:

consist of pairs of data ($Y_i,x_i$), where $Y$ is a random variable measured at each data location, e.g. rainfall, temperature. The task is usually to describe the relationship between the $Y_i$, allowing for their spatial locations.

Lattice data:

are similar to geostatistical data, but are located on a ‘lattice’ or ‘graph’ structure, e.g. crop yield in grid cells covering a field.

We will look at each of these types of spatial data during the course.