Scribd
Upload
44 views

13 - Statistical Analysis

This document discusses spatial statistics and how they are used in ArcGIS. It defines spatial statistics as analytical methods that incorporate spatial relationships to describe and analyze spatial patterns. It describes exploratory, inferential, and predictive spatial statistical methods. Specifically, it discusses how ArcGIS can be used to calculate measures of central tendency and dispersion, directional distribution, average nearest neighbor analysis, ordinary least squares regression, and geographically weighted regression. These statistical techniques help analyze and model relationships between spatial features and phenomena.

Uploaded by

dbecinho
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
44 views

13 - Statistical Analysis

This document discusses spatial statistics and how they are used in ArcGIS. It defines spatial statistics as analytical methods that incorporate spatial relationships to describe and analyze spatial patterns. It describes exploratory, inferential, and predictive spatial statistical methods. Specifically, it discusses how ArcGIS can be used to calculate measures of central tendency and dispersion, directional distribution, average nearest neighbor analysis, ordinary least squares regression, and geographically weighted regression. These statistical techniques help analyze and model relationships between spatial features and phenomena.

Uploaded by

dbecinho
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

GI Systems and Science

March 14, 2012


Points to Cover
 Conceptualizing spatial patterns
 Spatial statistics
 Definition
 Typology
 Spatial statistics in ArcGIS
Conceptualizing Spatial Patterns
 Geography is concerned with the study of
spatial patterns produced by physical and
cultural processes
 Description of the patterns
 Analysis of the patterns
 The spatial patterns can be conceptualized as
 Distribution of the features of phenomena across
the study area
 Distribution of cities in a region
 Distribution of a certain characteristic (variables)
associated with those features or phenomena
 Distribution of average income across the cities
Spatial Statistics
 Spatial statistical analytical methods are
tools used to describe and analyze spatial
patterns
 Allow ideas about spatial processes and patterns
to be tested
 Incorporate space (proximity, area, connectivity,
and/or other spatial relationships) directly into
their mathematics
 Are based on the assumption of spatial
autocorrelation
 Tobler’s First Law of Geography
Spatial Statistics
 Types of quantitative spatial statistical methods
 Exploratory and descriptive statistics
 Describe distribution of the spatial features or
phenomena
 Inferential statistics
 Compare sample pattern to a theoretical pattern of
distribution of features or phenomena
 Predictive statistics
 Analyse relationships between features or phenomena
 Prescriptive statistics
 Modeling and optimization methods
Spatial Statistics in GIS
 Exploratory and descriptive spatial statistics
 Simple summary statistics
 Histograms
 Measuring Geographic Distribution Toolset
 Spatial measures of central tendency
 Spatial measures of dispersion
Spatial Statistics in GIS Source: ArcGIS 10 Help files

 Directional Distribution
(Standard Deviational Ellipse)
 Measures a trend for a set of points or areas by
calculating the standard distance separately in the x-
and y-directions
Where
• Xi and Yi are coordinates
of a feature I
• X, Y are coordinates of
the mean centre for the
set of features
• n is the number of
features in the set
Spatial Statistics in GIS Source: ArcGIS 10 Help files

 Directional Distribution
(Standard Deviational Ellipse)
 Can be calculated using
 The locations of the features or
 The locations influenced by an attribute value
associated with the features
 Application
 Mapping the distributional trend for a set of crimes to
explore a potential relationship between incidents and
location of features
 Assessing the spread of contaminant based on the
contaminant concentrations collected at sample sites
 Comparing patterns over time
Spatial Statistics in GIS
 Inferential statistics
 Identifying geographic patterns is important for
understanding how geographic phenomena behave
 Analysing Patterns Toolset
○ Based on the null hypothesis that the features (or the
values associated with the features) exhibit a spatially
random pattern
○ Compute the probability that the null hypothesis is
correct
Spatial Statistics in GIS
 Average Nearest Neighbour Index
 Is expressed as the ratio of the observed mean
distance to the expected mean distance between
the features
 The expected distance is the average distance between
neighbours in a hypothetical random distribution
Where
• D0 is the observed mean
distance between a feature
and its nearest neighbour
• DE is the expected mean
distance between the features
given the random pattern
• A is the study area
Spatial Statistics in GIS
 Average Nearest Neighbour Index
 Based on the assumption that the points are free to
locate anywhere within the study area
 There are no barriers
 Features are located independently of one another

 Interpretation

ANN > 1 ANN = 1 ANN < 1


Spatial Statistics in GIS
 Average Nearest Neighbour Index
 Application
○ Quantify and compare the spatial distribution of a
variety of plant or animal species within a fixed study
area
 Monitor changes over time in the patterns of spatial
distribution of plant and animal species
 Compare an observed distribution to a control
distribution
Spatial Statistics in GIS
 Predictive Statistics
 Beyond analyzing spatial patterns, GIS analysis can be
used to examine or quantify relationships among
features
 The Modeling Spatial Relationships Toolset
 Construct spatial weights matrices
 Model spatial relationships using regression analyses
Spatial Statistics in GIS
 Ordinary Least Squares Regression
 Regression is used to evaluate relationships between
two or more feature attributes
 Helps to better understand processes active in an area
 Helps to predict where something is likely to occur
 Serves a start for exploring causal relationships
between location and characteristics of features
 OLS regression provides a global model of the
variable or process
 A good starting point for more sophisticated regression
analyses
Spatial Statistics in GIS
 Ordinary Least Squares Regression
 Regression is used to evaluate relationships between
two or more feature attributes
 Helps to better understand processes active in an area
 Helps to predict where something is likely to occur
 Serves as a start for exploring causal relationships
between location and characteristics of features
 OLS regression provides a global model of the
variable or process
 Creates a single regression model describing a process
 A good starting point for more sophisticated regression
analyses
Spatial Statistics in GIS
 Ordinary Least Squares Regression
 Issues with applying OLS to spatial data
 Spatial autocorrelation creates an over count type of
bias
 Nonstationary nature of many geographic processes
 Regional variation
 Processes may behave differently in different parts of the area
 Geographically weighted regression (GWR)
 Provides a local model of the variable or process by
fitting a regression equation to every feature in the
dataset
 Deal effectively with the spatial autocorrelation issue

You might also like