Exploratory spatial data analysis

8 important questions on Exploratory spatial data analysis

What is the definition of the spatial system (Dublin 2009) (step 1)?

Spatial neighborhood matrix W
n x n dimension
W is commonly sparse --> only a small number of elements are neighbors
Statistics are sensitive concering a selected neighborhood definition

What is the definition of a contiguity (step 1)?

Is often used for lattice data
Binary weight matrix: 1 if they share a boundry, 0 otherwise
Contiguity most appropriate for areal units 
Topologically one can consider nodes and/or edges as boundary (Rooks Case/Queens Case)

What is the definition of k-nearest neighbors (step 1)?

Distance based definitions more suited for point data
k closest entities are defined as neighbors
Avoids "island effects"
Both require plane and projected coordinates --> Euclidean distance
For latitude longitude --> arc distance/great circle distance
Suitable when units vary in size --> similar number of neighbors
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

What is the definition of an interaction (step 1)?

Spatial closeness results in similarity
Closer entities have greater influence than more distant ones
Common functions: inverse distance and squared inverse distance

What is the permutation approach (step 3)?

Is an alternative without making distributional assumptions
How unusual is our observed pattern?
When we run the Moran's I 999 times --> reference distribution --> if the Moran's I lies in a tail --> statistical evidence against chane --> reject H0

p-value: represents the probability of obtaining a test statistic at least as extreme as the observed one (usually 0,05)
if p < 0,05 --> reject H0 and conclude that similar values are spatially grouped

What are advantages of Local statistics?

Output of many parameters
Visualization capabilities
Detection of clusters
Explore hetrogeneity

What is the Local Moran's I?

The determination of attribute similarity for each unit in comparison to its neighborhood
Sum of local coefficients proportional to the global index

Hot or cold spots
Spatial outliers

Explain the moran scatterplot (step 3, local)

  • Slope of the regression line corresponds to the global Moran's I


High-high: hot spots --> positive SAC
low-low: cold spot --> positive SAC
High-low: outliers --> negative SAC
low-high: outliers --> negative SAC

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo