Introduction

Visualizing spatial data

Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial Data Analysis in Stata

An Overview
Maurizio Pisati
Department of Sociology and Social Research
University of Milano-Bicocca (Italy)
[email protected]

2012 Italian Stata Users Group meeting

Bologna
September 20-21, 2012
Maurizio Pisati

Spatial Data Analysis in Stata

1/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Outline
1

Introduction
Spatial data analysis in Stata
Space, spatial objects, spatial data

Maurizio Pisati

Spatial Data Analysis in Stata

2/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Outline
1

Introduction
Spatial data analysis in Stata
Space, spatial objects, spatial data

Visualizing spatial data

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Maurizio Pisati

Spatial Data Analysis in Stata

2/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Outline
1

Introduction
Spatial data analysis in Stata
Space, spatial objects, spatial data

Visualizing spatial data

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Exploring spatial point patterns

Overview
Kernel density estimation
Maurizio Pisati

Spatial Data Analysis in Stata

2/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Outline

Measuring spatial proximity

Maurizio Pisati

Spatial Data Analysis in Stata

3/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Outline

Measuring spatial proximity

Detecting spatial autocorrelation

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Maurizio Pisati

Spatial Data Analysis in Stata

3/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Outline

Measuring spatial proximity

Detecting spatial autocorrelation

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Fitting spatial regression models

Maurizio Pisati

Spatial Data Analysis in Stata

3/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Introduction

Maurizio Pisati

Spatial Data Analysis in Stata

4/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Spatial data analysis in Stata

Stata users can perform spatial data analysis using a

variety of user-written commands published in the Stata

Technical Bulletin, the Stata Journal, or the SSC Archive

Maurizio Pisati

Spatial Data Analysis in Stata

5/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Spatial data analysis in Stata

Stata users can perform spatial data analysis using a

variety of user-written commands published in the Stata

Technical Bulletin, the Stata Journal, or the SSC Archive
In this talk, I will briefly illustrate the use of six such

commands: spmap, spgrid, spkde, spatwmat, spatgsa,

and spatlsa

Maurizio Pisati

Spatial Data Analysis in Stata

5/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Spatial data analysis in Stata

Stata users can perform spatial data analysis using a

variety of user-written commands published in the Stata

Technical Bulletin, the Stata Journal, or the SSC Archive
In this talk, I will briefly illustrate the use of six such

commands: spmap, spgrid, spkde, spatwmat, spatgsa,

and spatlsa
I will also mention a pair of Stata commands/suites for

fitting spatial regression models: spatreg and sppack

Maurizio Pisati

Spatial Data Analysis in Stata

5/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Spatial data: a discrete view

For simplicity, let us represent space as a plane, i.e., as a

flat two-dimensional surface

Maurizio Pisati

Spatial Data Analysis in Stata

6/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Spatial data: a discrete view

For simplicity, let us represent space as a plane, i.e., as a

flat two-dimensional surface

In spatial data analysis, we can distinguish two conceptions
of space (Bailey and Gatrell 1995: 18):
Entity view : Space as an area filled with a set of discrete

objects
Field view : Space as an area covered with essentially

continuous surfaces

Maurizio Pisati

Spatial Data Analysis in Stata

6/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Spatial data: a discrete view

For simplicity, let us represent space as a plane, i.e., as a

flat two-dimensional surface

In spatial data analysis, we can distinguish two conceptions
of space (Bailey and Gatrell 1995: 18):
Entity view : Space as an area filled with a set of discrete

objects
Field view : Space as an area covered with essentially

continuous surfaces
Here we take the former view and define spatial data as

information regarding a given set of discrete spatial objects

located within a study area A

Maurizio Pisati

Spatial Data Analysis in Stata

6/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Attributes of spatial objects

Information about spatial objects can be classified into two

categories:
Spatial attributes
Non-spatial attributes

Maurizio Pisati

Spatial Data Analysis in Stata

7/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Attributes of spatial objects

Information about spatial objects can be classified into two

categories:
Spatial attributes
Non-spatial attributes

The spatial attributes of a spatial object consist of one

or more pairs of coordinates that represent its shape

and/or its location within the study area

Maurizio Pisati

Spatial Data Analysis in Stata

7/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Attributes of spatial objects

Information about spatial objects can be classified into two

categories:
Spatial attributes
Non-spatial attributes

The spatial attributes of a spatial object consist of one

or more pairs of coordinates that represent its shape

and/or its location within the study area
The non-spatial attributes of a spatial object consist of

its additional features that are relevant to the analysis at

hand

Maurizio Pisati

Spatial Data Analysis in Stata

7/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Types of spatial objects

According to their spatial attributes, spatial objects can

be classified into several types

Maurizio Pisati

Spatial Data Analysis in Stata

8/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Types of spatial objects

According to their spatial attributes, spatial objects can

be classified into several types

Here, we focus on two basic types:
Points (point data)
Polygons (area data)

Maurizio Pisati

Spatial Data Analysis in Stata

8/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Points

A point si is a zero-dimensional

spatial object located within

study area A at coordinates
(si1 , si2 )

Maurizio Pisati

Spatial Data Analysis in Stata

9/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Points

A point si is a zero-dimensional

spatial object located within

study area A at coordinates
(si1 , si2 )
Points can represent several kinds

of real entities, e.g., dwellings,

buildings, places where specific
events took place, pollution
sources, trees

Maurizio Pisati

Spatial Data Analysis in Stata

9/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Points

A point si is a zero-dimensional

spatial object located within

study area A at coordinates
(si1 , si2 )
Points can represent several kinds

of real entities, e.g., dwellings,

buildings, places where specific
events took place, pollution
sources, trees

Maurizio Pisati

Spatial Data Analysis in Stata

9/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Points
Homicides
Washington D.C. (2009)

A point si is a zero-dimensional

spatial object located within

study area A at coordinates
(si1 , si2 )
Points can represent several kinds

of real entities, e.g., dwellings,

buildings, places where specific
events took place, pollution
sources, trees

Maurizio Pisati

Spatial Data Analysis in Stata

9/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Polygons
A polygon ri is a region of study

area A bounded by a closed

polygonal chain whose M 4
vertices are defined by the
coordinate set {(ri1(1) , ri2(1) ),
(ri1(2) , ri2(2) ), . . . , (ri1(m) , ri2(m) ),
. . . , (ri1(M ) , ri2(M ) )}, where
ri1(1) = ri1(M ) and ri2(1) = ri2(M )

Maurizio Pisati

Spatial Data Analysis in Stata

10/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Polygons
A polygon ri is a region of study

area A bounded by a closed

polygonal chain whose M 4
vertices are defined by the
coordinate set {(ri1(1) , ri2(1) ),
(ri1(2) , ri2(2) ), . . . , (ri1(m) , ri2(m) ),
. . . , (ri1(M ) , ri2(M ) )}, where
ri1(1) = ri1(M ) and ri2(1) = ri2(M )
Polygons can represent several

kinds of real entities, e.g., states,

provinces, counties, census tracts,
electoral districts, parks, lakes
Maurizio Pisati

Spatial Data Analysis in Stata

10/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Polygons
A polygon ri is a region of study

area A bounded by a closed

polygonal chain whose M 4
vertices are defined by the
coordinate set {(ri1(1) , ri2(1) ),
(ri1(2) , ri2(2) ), . . . , (ri1(m) , ri2(m) ),
. . . , (ri1(M ) , ri2(M ) )}, where
ri1(1) = ri1(M ) and ri2(1) = ri2(M )
Polygons can represent several

kinds of real entities, e.g., states,

provinces, counties, census tracts,
electoral districts, parks, lakes
Maurizio Pisati

Spatial Data Analysis in Stata

10/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial data analysis in Stata

Space, spatial objects, spatial data

Polygons
A polygon ri is a region of study

area A bounded by a closed

polygonal chain whose M 4
vertices are defined by the
coordinate set {(ri1(1) , ri2(1) ),
(ri1(2) , ri2(2) ), . . . , (ri1(m) , ri2(m) ),
. . . , (ri1(M ) , ri2(M ) )}, where
ri1(1) = ri1(M ) and ri2(1) = ri2(M )

Police Districts
Washington D.C.

Fourth

Second
Third

Fifth

First

Sixth

Polygons can represent several

kinds of real entities, e.g., states,

provinces, counties, census tracts,
electoral districts, parks, lakes
Maurizio Pisati

Seventh

Spatial Data Analysis in Stata

10/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Visualizing spatial data

Maurizio Pisati

Spatial Data Analysis in Stata

11/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Thematic maps
Most analyses of spatial data have their natural starting

point in displaying the information of interest by one or

more maps
If properly designed, maps can help the analyst to detect

interesting patterns in the data, spatial relationships

between two or more phenomena, unusual observations,
and so on
Thematic maps represent the spatial distribution of a

phenomenon of interest within a given study area (Slocum

et al. 2005)

Maurizio Pisati

Spatial Data Analysis in Stata

12/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Thematic maps in Stata

Stata users can generate thematic maps using spmap, a

user-written command freely available from the SSC

Archive (latest version: 1.2.0)
spmap is a very flexible command that allows for creating a

large variety of thematic maps, from the simplest to the

most complex
While providing sensible defaults for most options and

supoptions, spmap gives the user full control over the

formatting of almost every map element, thus allowing the
production of highly customized maps

Maurizio Pisati

Spatial Data Analysis in Stata

13/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Thematic maps in Stata

In the following, I will show some examples on using spmap

for creating common types of thematic maps:

Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Maurizio Pisati

Spatial Data Analysis in Stata

14/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Dot maps

A dot map shows the spatial distribution of a set of point

spatial objects S {si ; i = 1, . . . , N }, i.e., their location

within a given study area A
If the point spatial objects have variable attributes, it is

possible to represent this information using symbols of

different colors and/or of different shape

Maurizio Pisati

Spatial Data Analysis in Stata

15/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Dot maps: example 1

Spatial distribution of 359 cases of sex

abuse, Washington D.C. (2009). Different
colors are used to distinguish adult
victims from child victims

Sex abuses, by victim age

Washington D.C. (2009)

!"#$%&'()#*++,-./0%!$12#0'$
3#4#'0/#$567$"$#$$
3#4#'0/#$8(1/()$"$)#/9:.$
'#1:.#$8(1/()$%&'(")*%)(')+",*%-".*$
20;#2$.#<(4#$8(1/()$)$%=.!2/%$,$%&9(2.%$
20;#2$802!#"$8(1/()$8(1/()$
">)0>$!"(43$%?:!4.0'(#"-./0%!$(.%567*$<1:2:'%#33"9#22*$$$$$$'''$
$$$>:(4/%@%@51::'.*$A%A51::'.*$"#2#1/%B##>$(<$:<<#4"#""/*$$$'''$
$$$$$;A%8(1/()*$"(C#%-).,*$<1:2:'%'#.$408A*$$$$$$$$$$$$$$$$$'''$
$$$$$:1:2:'%D9(/#$..*$:"(C#%-0.1$..*$2#3#4.0%:4*$$$$$$$$$$$$'''$
$$$$$2#31:!4/*$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$'''$
$$$2#3#4.%"(C#%-).+*$':D30>%).,**$$$$$$$$$$$$$$$$$$$$$$$$$$$'''$
$$$/(/2#%%E#@$0;!"#"F$;A$8(1/()$03#%!$"(C#%-).,**$$$$$$$$$$$'''$
$$$"!;/(/2#%%G0"9(43/:4$7-&-$H*++,I%$%$%!$"(C#%-).,**!

Adult (155)
Child (204)

Maurizio Pisati

Spatial Data Analysis in Stata

16/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Dot maps: example 2

Sex abuses
Washington D.C. (2009)

Spatial distribution of 359 cases of sex

abuse, Washington D.C. (2009). Major
roads, watercourses and parks are added
to the map for reference
!"#$%&'()#*++,-./0%!$12#0'$
3#4#'0/#$567$"$#$$
"8)08$!"(43$%9:!4.0'(#"-./0%!$(.%567&$;1:2:'%#33"<#22&$$$$$$'''$
$$$8:(4/%=%=51::'.&$>%>51::'.&$"#2#1/%?##8$(;$:;;#4"#""(&$$$'''$
$$$$$"(@#%)*+,&$;1:2:'%'#.&$:1:2:'%A<(/#&$:"(@#%)-+.&&$$$$$$'''$
$$$8:2>3:4%.0/0%%B0/#'CD0'?"-./0%&$E>%/>8#&$$$$$$$$$$$$$$$$$'''$
$$$$$:1:2:'%4:4#$++&$;1:2:'%3'##4$E2!#&&$$$$$$$$$$$$$$$$$$$$'''$
$$$2(4#%.0/0%%F0G:'H:0."-./0%&$1:2:'%E':A4&&$$$$$$$$$$$$$$$$'''$
$$$/(/2#%%I#=$0E!"#"%!$"(@#%)*+,&&$$$$$$$$$$$$$$$$$$$$$$$$$$'''$
$$$"!E/(/2#%%B0"<(43/:4$7-&-$J*++,K%$%$%!$"(@#%)*+,&&!

Maurizio Pisati

Spatial Data Analysis in Stata

17/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Proportional symbol maps

A proportional symbol map represents the values taken

by a numeric variable of interest Y on a set of point spatial

objects S located within a given study area A
Proportional symbol maps can be used with two types of
point data (Slocum et al. 2005: 310):
True point data are measured at actual point locations
Conceptual point data are collected over a set of regions

R {ri ; i = 1, . . . , N }, but are conceived as being located

at representative points within the regions, typically at
their centroids
The area of each point symbol is sized in direct proportion

to the corresponding value of Y

Maurizio Pisati

Spatial Data Analysis in Stata

18/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Proportional symbol maps: example

Mean family income (in thousands of US dollars)
Washington D.C. (2000)

Mean family income in the seven Police

Districts of Washington D.C. (2000)
19.6

!"#$%&'()*#+)",-)*,".+/,/01,/%!$*(#/-$
2#3#-/,#$4$"$)3*'5#65/#$%%%$
7'-5/,$4$&'($7$
"85/8$!")32$%&'()*#+)",-)*,".9''-1)3/,#"01,/%!$)1))1*$$$$$$$###$
$$$7*'('-)#22":#((*$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$8')3,););6*''-1*$<)<6*''-1*$8-'8'-,)'3/()4*$7*'('-)-#1*$$###$
$$$$$'*'('-)=:),#*$")>#)+,(-**$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$(/?#(););6*''-1*$<*''-1)<6*''-1*$(/?#()4*$*'('-)=:),#*$$$###$
$$$$$")>#)+$('**$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$,),(#)%@#/3$7/5)(<$)3*'5#$A)3$,:'!"/31"$'7$BC$1'((/-"D%*$###$
$$$"!?,),(#)%E/":)32,'3$+090$AFGGGD%$%$%*!

51.5
22.7

14.0

26.1

12.7

8.1

Maurizio Pisati

Spatial Data Analysis in Stata

19/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Diagram maps

A diagram map follows the same logic as a proportional

symbol map, but represents the values of the variable of

interest using bar charts, pie charts, or other types of
diagram
The use of pie charts allows to display the spatial

distribution of compositional data, i.e., of two or more

numeric variables that represent parts of a whole

Maurizio Pisati

Spatial Data Analysis in Stata

20/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Diagram maps: example 1

Mean family income
Washington D.C. (2000)

Mean family income in the seven Police

Districts of Washington D.C. (2000).
Data are represented by framed-rectangle
charts, with the overall mean income as
the reference value
!"#$%&'()*#+)",-)*,".+/,/01,/%!$*(#/-$
"23/2$!")45$%&'()*#+)",-)*,".6''-1)4/,#"01,/%!$)1")1#$$$$$$$$$$$
$$$7*'('-"#55"8#((#$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$$$1)/5-/3"9/-")4*'3#:3/#$-#7;#)58,"2'2,',#$7*'('-"5-##4#$$$$$$$
$$$$$$<"<:*''-1#$="=:*''-1#$")>#"%&'##$$$$$$$$$$$$$$$$$$$$$$$$$$
$$$,),(#"%?#/4$7/3)(=$)4*'3#%#$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$$$"!@,),(#"%A/"8)45,'4$+060$BCDDDE%$%$%#!

Maurizio Pisati

Spatial Data Analysis in Stata

21/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Diagram maps: example 2

Race/Ethnic composition of the
population of the seven Police Districts of
Washington D.C. (2000). Data are
represented by pie charts

Race/Ethnic composition of the population

Washington D.C. (2000)

!"#$%&'()*#+)",-)*,".+/,/01,/%!$*(#/-$
2#3#-/,#$45),#67*,$"$7'7645),##7'7,',$%&&$
2#3#-/,#$/8-'/967*,$"$7'76/8-'/9#7'7,',$%&&$
2#3#-/,#$',5#-67*,$"$7'76',5#-#7'7,',$%&&$
(/:#($;/-)/:(#$45),#67*,$%<5),#%$
(/:#($;/-)/:(#$/8-'/967*,$%=8-)*/3$=9#-)*/3%$
(/:#($;/-)/:(#$',5#-67*,$%>,5#-%$
"79/7$!")32$%&'()*#+)",-)*,".?''-1)3/,#"01,/%!$)1')1($$$$$$$###$
$$$8*'('-'",'3#($$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$1)/2-/9';/-'45),#67*,$/8-'/967*,$',5#-67*,($@'@6*''-1($$$###$
$$$$$A'A6*''-1($8*'('-'#22"5#(($-#1$'-/32#($")B#'%)*($$$$$$$###$
$$$$$(#2#31/''3(($$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$(#2#31'")B#'$%)+(($$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$,),(#'%C/*#DE,53)*$*'97'"),)'3$'8$,5#$7'7!(/,)'3%($$$$$$$###$
$$$"!:,),(#'%</"5)32,'3$+0?0$FGHHHI%$%$%(!
White
African American
Other

Maurizio Pisati

Spatial Data Analysis in Stata

22/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Choropleth maps

A choropleth map displays the values taken by a variable

of interest Y on a set of regions R within a given study

area A
When Y is numeric, each region is colored or shaded

according to a discrete scale based on its value on Y

The number of classes k that make up the discrete scale,

and the corresponding class breaks, can be based on several

different criteria e.g., quantiles, equal intervals, boxplot,
standard deviates

Maurizio Pisati

Spatial Data Analysis in Stata

23/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Choropleth maps: example 1

Mean family income (in thousands of US dollars)
Washington D.C. (2000)

Mean family income in the 188 Census

Tracts of Washington D.C. (2000).
Income is divided into six classes based
on the quantiles method
!"#$%&#'"!"()))*+,-,./-,%!$01#,2$
3#'#2,-#$4$"$5'067#87,#$%%%$
9627,-$4$&'(%9$
":7,:$4$!"5'3$%&#'"!"()))*&662/5',-#"./-,%!$5/)5/*$$$$$$$$$$###$
$$$01'!7;#2)+*$017#-<6/)=!,'-51#*$906162)>!?/*$$$$$$$$$$$$$$###$
$$$'/906162)3"@*$'/1,;)%A5""5'3%*$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$1#3#'/)"5B#),$(-**$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$-5-1#)%A#,'$9,751C$5'067#$D5'$-<6!",'/"$69$EF$/611,2"G%*$###$
$$$"!;-5-1#)%H,"<5'3-6'$+.&.$D()))G%$%$%*!
(93,341]
(54,93]
(40,54]
(33,40]
(28,33]
[6,28]
Missing

Maurizio Pisati

Spatial Data Analysis in Stata

24/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Choropleth maps: example 2

Mean family income (in thousands of US dollars)
Washington D.C. (2000)

Mean family income in the 188 Census

Tracts of Washington D.C. (2000).
Income is divided into six classes based
on the boxplot method
!"#$%&#'"!"()))*+,-,./-,%!$01#,2$
3#'#2,-#$4$"$5'067#87,#$%%%$
9627,-$4$&'(%9$
":7,:$4$!"5'3$%&#'"!"()))*&662/5',-#"./-,%!$5/)5/*$$$$$$$$$$###$
$$$01'!7;#2)+*$017#-<6/);6=:16-*$906162)>!?/*$$$$$$$$$$$$$$$###$
$$$'/906162)3"@*$'/1,;)%A5""5'3%*$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$1#3#'/)"5B#),$(-**$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$-5-1#)%A#,'$9,751C$5'067#$D5'$-<6!",'/"$69$EF$/611,2"G%*$###$
$$$"!;-5-1#)%H,"<5'3-6'$+.&.$D()))G%$%$%*!
(132,341]
(71,132]
(40,71]
(31,40]
(6,31]
[6,6]
Missing

Maurizio Pisati

Spatial Data Analysis in Stata

25/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Multivariate maps

A multivariate map combines several types of thematic

mapping to simultaneously display the spatial distribution

of multiple phenomena within a given study area A

Maurizio Pisati

Spatial Data Analysis in Stata

26/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Dot maps
Proportional symbol maps
Diagram maps
Choropleth maps
Multivariate maps

Multivariate maps: example

The map shows the relationship between
pct. white population (represented by
framed-rectangle charts), mean family
income (represented by the width of
framed-rectangle charts) and robbery
rate (represented by shades of color)
across the seven Police Districts of
Washington D.C. (2000/2009)
!"#$%&'()*#+)",-)*,".+/,/01,/%!$*(#/-$
2#3#-/,#$4$"$5'5678),##5'5,',$%&&$
9'-:/,$4$'()&9$
"5:/5$4$!")32$%&'()*#+)",-)*,".;''-1)3/,#"01,/%!$)1*)1+$$$$$###$
$$$*(:#,8'1**!",':+$*(<-#/="*&$(,$,&$-,$%&&+$9*'('-*4(>3+$$$###$
$$$(#2,),*%&*,0$78),#$5'5!(/,)'3%+$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$1)/2-/:*?/-*)3*':#6:/+$-#97#)28,*5'5,',+$9*'('-*-#1+$$$$$###$
$$$$$@*@6*''-1+$A*A6*''-1+$")B#*%).++$$$$$$$$$$$$$$$$$$$$$$$###$
$$$(#2#31*")B#*$%)/++$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$###$
$$$,),(#*%C#/3$9/:)(A$)3*':#$/31$5*,0$78),#$5'5!(/,)'3%+$$$$###$
$$$"!<,),(#*%D/"8)32,'3$+0;0$EFGGGH%$%$%+!

Maurizio Pisati

Pct. white population, income and robberies

Washington D.C. (2000/2009)

Robberies per 1,000 pop.

(10,12]
(8,10]
(6,8]
[2,6]

Spatial Data Analysis in Stata

27/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Exploring spatial point patterns

Maurizio Pisati

Spatial Data Analysis in Stata

28/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Two-dimensional spatial point patterns

A two-dimensional spatial point pattern can be

defined as a set of N point spatial objects S located within

a given study area A
Usually, each point si S represents a real entity of some

kind: people, events, sites, buildings, plants, cases of a

disease, etc.
Alternatively, each point si represents the centroid of a

region
Points si are referred to as the data points

Maurizio Pisati

Spatial Data Analysis in Stata

29/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Two-dimensional spatial point patterns

In the analysis of spatial point patterns, we are often

interested in determining whether the observed data points

exhibit some form of clustering, as opposed to being
distributed uniformly within A

Maurizio Pisati

Spatial Data Analysis in Stata

30/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Two-dimensional spatial point patterns

In the analysis of spatial point patterns, we are often

interested in determining whether the observed data points

exhibit some form of clustering, as opposed to being
distributed uniformly within A
To explore the possibility of point clustering, it may be

useful to describe the spatial point pattern of interest by

means of its probability density function p(s) and/or its
intensity function (s) (Waller and Gotway 2004)

Maurizio Pisati

Spatial Data Analysis in Stata

30/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Two-dimensional spatial point patterns

The probability density function p(s) defines the

probability of observing an object per unit area at location

sA
The intensity function (s) defines the expected number

of objects per unit area at location s A

The probability density function and the intensity function

differ only by a constant of proportionality

Maurizio Pisati

Spatial Data Analysis in Stata

31/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimators
Both the probability density function p(s) and the intensity

function (s) of a two-dimensional spatial point pattern

can be estimated by means of nonparametric estimators,
e.g., kernel estimators (Waller and Gotway 2004)

Maurizio Pisati

Spatial Data Analysis in Stata

32/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimators
Both the probability density function p(s) and the intensity

function (s) of a two-dimensional spatial point pattern

can be estimated by means of nonparametric estimators,
e.g., kernel estimators (Waller and Gotway 2004)
Kernel estimators are used to generate a spatially

smooth estimate of p(s) and/or (s) at a fine grid of points

sg (g = 1, ..., G) covering the study area A

Maurizio Pisati

Spatial Data Analysis in Stata

32/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimators
Both the probability density function p(s) and the intensity

function (s) of a two-dimensional spatial point pattern

can be estimated by means of nonparametric estimators,
e.g., kernel estimators (Waller and Gotway 2004)
Kernel estimators are used to generate a spatially

smooth estimate of p(s) and/or (s) at a fine grid of points

sg (g = 1, ..., G) covering the study area A
In the context of spatial data analysis, a grid is a regular

tessellation of the study area A that divides it into a set of

G contiguous cells whose centers are referred to as the grid
points and denoted by sg
Maurizio Pisati

Spatial Data Analysis in Stata

32/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation in Stata

Stata users can generate kernel estimates of the probability

density function p(s) and the intensity function (s) using

two user-written commands freely available from the SSC
Archive: spgrid and spkde

Maurizio Pisati

Spatial Data Analysis in Stata

33/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation in Stata

Stata users can generate kernel estimates of the probability

density function p(s) and the intensity function (s) using

two user-written commands freely available from the SSC
Archive: spgrid and spkde
spgrid (latest version: 1.0.1) generates several kinds of

two-dimensional grids covering rectangular or irregular

study areas

Maurizio Pisati

Spatial Data Analysis in Stata

33/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation in Stata

Stata users can generate kernel estimates of the probability

density function p(s) and the intensity function (s) using

two user-written commands freely available from the SSC
Archive: spgrid and spkde
spgrid (latest version: 1.0.1) generates several kinds of

two-dimensional grids covering rectangular or irregular

study areas
spkde (latest version: 1.0.0) implements a variety of kernel

estimators of p(s) and (s)

Maurizio Pisati

Spatial Data Analysis in Stata

33/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation in Stata

Stata users can generate kernel estimates of the probability

density function p(s) and the intensity function (s) using

two user-written commands freely available from the SSC
Archive: spgrid and spkde
spgrid (latest version: 1.0.1) generates several kinds of

two-dimensional grids covering rectangular or irregular

study areas
spkde (latest version: 1.0.0) implements a variety of kernel

estimators of p(s) and (s)

spmap can then be used to visualize the kernel estimates

generated by spgrid and spkde

Maurizio Pisati

Spatial Data Analysis in Stata

33/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation: example

Our purpose is to estimate the probability density function of

a set of 139 points representing the homicides committed in
Washington D.C. in 2009

Maurizio Pisati

Spatial Data Analysis in Stata

34/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation: example

Step 1
We use spgrid to generate a grid
covering the area of Washington D.C. We
choose a relatively fine grid resolution
(grid cell width = 200 meters). spmap is
used to display the grid

Maurizio Pisati

Spatial Data Analysis in Stata

35/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation: example

Step 1
We use spgrid to generate a grid
covering the area of Washington D.C. We
choose a relatively fine grid resolution
(grid cell width = 200 meters). spmap is
used to display the grid
!"#$%&'(!%)#'*+,()&-$%.!/&0-*!'$.!,1(0%,)"2344#''''$$$'
'''&,0!'5,6"$.!!'()%0"6.0.$!#'5.11!"*50.6"/&0-*#'''$$$'
'''",%)0!"*"0.6"/&0-*#'$."1-5.'
'
(!.'*"0.6"/&0-*!'51.-$'
!"6-"'(!%)#'*50.6"/&0-*!'%&"!"#$%&7%&#!

Maurizio Pisati

Spatial Data Analysis in Stata

35/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation: example

Step 1
We use spgrid to generate a grid
covering the area of Washington D.C. We
choose a relatively fine grid resolution
(grid cell width = 200 meters). spmap is
used to display the grid
!"#$%&'(!%)#'*+,()&-$%.!/&0-*!'$.!,1(0%,)"2344#''''$$$'
'''&,0!'5,6"$.!!'()%0"6.0.$!#'5.11!"*50.6"/&0-*#'''$$$'
'''",%)0!"*"0.6"/&0-*#'$."1-5.'
'
(!.'*"0.6"/&0-*!'51.-$'
!"6-"'(!%)#'*50.6"/&0-*!'%&"!"#$%&7%&#!

Maurizio Pisati

Spatial Data Analysis in Stata

35/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation: example

Step 2
We use spkde to generate kernel
estimates of the probability distribution
of homicides in Washington D.C. We
choose a quartic kernel function with
fixed bandwidth equal to 1,000 meters
and edge correction. spmap is used to
display the results

Maurizio Pisati

Spatial Data Analysis in Stata

36/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation: example

Step 2
We use spkde to generate kernel
estimates of the probability distribution
of homicides in Washington D.C. We
choose a quartic kernel function with
fixed bandwidth equal to 1,000 meters
and edge correction. spmap is used to
display the results
!"#$%&'()#*++,-./0%!$12#0'$
3##4$(5$655#7"#""#$
"43.#$!"(78$%4/#)4-./0%!$9$9:166'.%$;$;:166'.%$$$&&&$
$$$3#'7#2$<!0'/(1%$=07.>(./?$5=>%$5=>$'(((%$$$$$$&&&$
$$$#.8#16''#1/$.6/"$"0@(78$%3.#-./0%!$'#4201#%$
$
!"#$%3.#-./0%!$12#0'$
"4)04$4$!"(78$%1/#)4-./0%!$(.$"48'(.:(.%$12)#/?6.$<!07/(2#%$&&&$
$$$127!)=#'$)(%$51626'$A0(7=6>%$61626'$767#$**%$2#8#7.$655%$&&&$
$$$/(/2#$%B6)(1(.#"%!$"(C#$+'*)%%$$$$$$$$$$$$$$$$$$$$$$$$$$$&&&$
$$$"!=/(/2#$%D0"?(78/67$E-&-$F*++,G%$%$%!$"(C#$+'*)%%!

Maurizio Pisati

Spatial Data Analysis in Stata

36/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation: example

Step 2
We use spkde to generate kernel
estimates of the probability distribution
of homicides in Washington D.C. We
choose a quartic kernel function with
fixed bandwidth equal to 1,000 meters
and edge correction. spmap is used to
display the results

Homicides
Washington D.C. (2009)

!"#$%&'()#*++,-./0%!$12#0'$
3##4$(5$655#7"#""#$
"43.#$!"(78$%4/#)4-./0%!$9$9:166'.%$;$;:166'.%$$$&&&$
$$$3#'7#2$<!0'/(1%$=07.>(./?$5=>%$5=>$'(((%$$$$$$&&&$
$$$#.8#16''#1/$.6/"$"0@(78$%3.#-./0%!$'#4201#%$
$
!"#$%3.#-./0%!$12#0'$
"4)04$4$!"(78$%1/#)4-./0%!$(.$"48'(.:(.%$12)#/?6.$<!07/(2#%$&&&$
$$$127!)=#'$)(%$51626'$A0(7=6>%$61626'$767#$**%$2#8#7.$655%$&&&$
$$$/(/2#$%B6)(1(.#"%!$"(C#$+'*)%%$$$$$$$$$$$$$$$$$$$$$$$$$$$&&&$
$$$"!=/(/2#$%D0"?(78/67$E-&-$F*++,G%$%$%!$"(C#$+'*)%%!

Maurizio Pisati

Spatial Data Analysis in Stata

36/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Kernel density estimation

Kernel estimation: example

Step 2
We use spkde to generate kernel
estimates of the probability distribution
of homicides in Washington D.C. We
choose a quartic kernel function with
fixed bandwidth equal to 1,000 meters
and edge correction. spmap is used to
display the results

Homicides
Washington D.C. (2009)

!"#$%&'()#*++,-./0%!$12#0'$
3##4$(5$655#7"#""#$
"43.#$!"(78$%4/#)4-./0%!$9$9:166'.%$;$;:166'.%$$$&&&$
$$$3#'7#2$<!0'/(1%$=07.>(./?$5=>%$5=>$'(((%$$$$$$&&&$
$$$#.8#16''#1/$.6/"$"0@(78$%3.#-./0%!$'#4201#%$
$
!"#$%3.#-./0%!$12#0'$
"4)04$4$!"(78$%1/#)4-./0%!$(.$"48'(.:(.%$12)#/?6.$<!07/(2#%$&&&$
$$$127!)=#'$)(%$51626'$A0(7=6>%$61626'$767#$**%$2#8#7.$655%$&&&$
$$$/(/2#$%B6)(1(.#"%!$"(C#$+'*)%%$$$$$$$$$$$$$$$$$$$$$$$$$$$&&&$
$$$"!=/(/2#$%D0"?(78/67$E-&-$F*++,G%$%$%!$"(C#$+'*)%%!

Maurizio Pisati

Spatial Data Analysis in Stata

37/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Measuring spatial proximity

Maurizio Pisati

Spatial Data Analysis in Stata

38/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial weights matrix

Most spatial data analyses require that the degree of

spatial proximity among the spatial objects of interest

be expressed in some way

Maurizio Pisati

Spatial Data Analysis in Stata

39/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial weights matrix

Most spatial data analyses require that the degree of

spatial proximity among the spatial objects of interest

be expressed in some way
Typically, the degree of spatial proximity among a given

set of N spatial objects is represented by a N N matrix

called spatial weights matrix and denoted by W

Maurizio Pisati

Spatial Data Analysis in Stata

39/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial weights matrix

Most spatial data analyses require that the degree of

spatial proximity among the spatial objects of interest

be expressed in some way
Typically, the degree of spatial proximity among a given

set of N spatial objects is represented by a N N matrix

called spatial weights matrix and denoted by W
Each element (i, j) of W which we denote by wij

expresses the degree of spatial proximity between the pair

of objects i and j

Maurizio Pisati

Spatial Data Analysis in Stata

39/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial weights matrix

Most spatial data analyses require that the degree of

spatial proximity among the spatial objects of interest

be expressed in some way
Typically, the degree of spatial proximity among a given

set of N spatial objects is represented by a N N matrix

called spatial weights matrix and denoted by W
Each element (i, j) of W which we denote by wij

expresses the degree of spatial proximity between the pair

of objects i and j
Depending on the application, the N main diagonal

elements of W are assigned value wii = 0 or value wii > 0

Maurizio Pisati

Spatial Data Analysis in Stata

39/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial weights matrix

A common variant of W is the row-standardized spatial

weights matrix Wstd , whose elements are defined as

follows:
std
wij
=

wij
N
P

wij

j=1

Maurizio Pisati

Spatial Data Analysis in Stata

40/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial weights matrices in Stata

Stata users can generate several kinds of spatial weights

matrices using spatwmat, a user-written command

published in the Stata Technical Bulletin (Pisati 2001)

Maurizio Pisati

Spatial Data Analysis in Stata

41/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial weights matrices in Stata

Stata users can generate several kinds of spatial weights

matrices using spatwmat, a user-written command

published in the Stata Technical Bulletin (Pisati 2001)
spatwmat (latest version: 1.0) imports or generates from

scratch the spatial weights matrices required by other

commands for spatial data analysis (see below)

Maurizio Pisati

Spatial Data Analysis in Stata

41/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Detecting spatial autocorrelation

Maurizio Pisati

Spatial Data Analysis in Stata

42/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Spatial autocorrelation
Forty years ago, the geographer and statistician Waldo

Tobler formulated the first law of geography: Everything

is related to everything else, but near things are more
related than distant things (Tobler 1970: 234)

Maurizio Pisati

Spatial Data Analysis in Stata

43/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Spatial autocorrelation
Forty years ago, the geographer and statistician Waldo

Tobler formulated the first law of geography: Everything

is related to everything else, but near things are more
related than distant things (Tobler 1970: 234)
This law defines the statistical concept of (positive)
spatial autocorrelation, according to which two or more
objects that are spatially close tend to be more similar to
each other with respect to a given attribute Y than are
spatially distant objects

Maurizio Pisati

Spatial Data Analysis in Stata

43/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Spatial autocorrelation
Forty years ago, the geographer and statistician Waldo

Tobler formulated the first law of geography: Everything

is related to everything else, but near things are more
related than distant things (Tobler 1970: 234)
This law defines the statistical concept of (positive)
spatial autocorrelation, according to which two or more
objects that are spatially close tend to be more similar to
each other with respect to a given attribute Y than are
spatially distant objects
In general, spatial autocorrelation implies spatial
custering, i.e., the existence of sub-areas of the study area
where the attribute of interest Y takes higher than average
values (hot spots) or lower than average values (cold spots)
Maurizio Pisati

Spatial Data Analysis in Stata

43/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Indices of spatial autocorrelation

We consider measures of spatial autocorrelation that apply

to area data

Maurizio Pisati

Spatial Data Analysis in Stata

44/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Indices of spatial autocorrelation

We consider measures of spatial autocorrelation that apply

to area data
Measures of spatial autocorrelation can be classified into
two broad categories:

Maurizio Pisati

Spatial Data Analysis in Stata

44/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Indices of spatial autocorrelation

We consider measures of spatial autocorrelation that apply

to area data
Measures of spatial autocorrelation can be classified into
two broad categories:
Global indices of spatial autocorrelation

Maurizio Pisati

Spatial Data Analysis in Stata

44/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Indices of spatial autocorrelation

We consider measures of spatial autocorrelation that apply

to area data
Measures of spatial autocorrelation can be classified into
two broad categories:
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Maurizio Pisati

Spatial Data Analysis in Stata

44/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Global indices of spatial autocorrelation

A global index of spatial autocorrelation expresses

the overall degree of similarity between spatially close

regions observed in a given study area A with respect to a
numeric variable Y (Pfeiffer et al. 2008)

Maurizio Pisati

Spatial Data Analysis in Stata

45/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Global indices of spatial autocorrelation

A global index of spatial autocorrelation expresses

the overall degree of similarity between spatially close

regions observed in a given study area A with respect to a
numeric variable Y (Pfeiffer et al. 2008)
Since global indices of spatial autocorrelation summarize

the phenomenon of interest in a single value, they are

intended not so much for identifying specific spatial
clusters, as for detecting the presence of a general tendency
to clustering within the study area

Maurizio Pisati

Spatial Data Analysis in Stata

45/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Global indices of spatial autocorrelation in Stata

Stata users can compute global indices of spatial auto-

correlation using spatgsa, a user-written command

published in the Stata Technical Bulletin (Pisati 2001)
spatgsa (latest version: 1.0) computes three global indices

of spatial autocorrelation: Morans I, Getis and Ords G,

and Gearys c. For each index and each numeric variable of
interest, spatgsa computes and displays in tabular form
the value of the index itself, the expected value of the index
under the null hypothesis of no global spatial autocorrelation, the standard deviation of the index, the
z -value, and the corresponding one- or two-tailed p-value

Maurizio Pisati

Spatial Data Analysis in Stata

46/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Global indices of spatial autocorrelation: example

Study area: Ohio

Regions: 88 counties
Variables of interest:
Pct. population aged 18+ with poor-to-fair health status
(pct poorhealth)
Pct. population aged 18+ currently smoking
(pct currsmoker)
Pct. population aged 18+ ever diagnosed with high blood
pressure (pct hibloodprs)
Pct. population aged 18+ obese (pct obese)

Maurizio Pisati

Spatial Data Analysis in Stata

47/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Global indices of spatial autocorrelation: example

Step 1
We use spatwmat to import an existing binary spatial weights matrix
stored in the Stata dataset Counties-Contiguity.dta and
convert it into a properly formatted row-standardized spatial weights
matrix Ws

Maurizio Pisati

Spatial Data Analysis in Stata

48/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Global indices of spatial autocorrelation: example

Step 1
We use spatwmat to import an existing binary spatial weights matrix
stored in the Stata dataset Counties-Contiguity.dta and
convert it into a properly formatted row-standardized spatial weights
matrix Ws

!"#$%&#$'(!)*+',-.(*$)/!0-.*$)+()$123$#,!'''"""'
'''*#&/#4!$'!$#*3#53)6/!

Maurizio Pisati

Spatial Data Analysis in Stata

48/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Global indices of spatial autocorrelation: example

r: Maurizio

The following matrix has been created:

1. Imported binary weights matrix Ws (row-standardized)
Dimension: 88x88

Maurizio Pisati

Spatial Data Analysis in Stata

49/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Global indices of spatial autocorrelation: example

Step 2
We use spatgsa with the spatial weights matrix Ws to compute
Morans I on the variables of interest

Maurizio Pisati

Spatial Data Analysis in Stata

50/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Global indices of spatial autocorrelation: example

Step 2
We use spatgsa with the spatial weights matrix Ws to compute
Morans I on the variables of interest
!"#$%&'!()*#"+,-)-./)-%!$01#-2$
"3-)4"-$30)53''26#-1)6$30)50!22"7'8#2$30)56*91''/32"$$$"""$
$$$30)5'9#"#!$:#;"$$7'2-(!

Maurizio Pisati

Spatial Data Analysis in Stata

50/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

User: Maurizio

Global indices of spatial autocorrelation: example

Measures of global spatial autocorrelation

Weights matrix
Name: Ws
Type: Imported (binary)
Row-standardized: Yes

Moran's I
Variables
pct_poorhealth
pct_currsmoker
pct_hibloodprs
pct_obese

I
0.399
0.339
0.126
0.167

E(I)

sd(I)

p-value*

-0.011
-0.011
-0.011
-0.011

0.065
0.065
0.065
0.065

6.337
5.367
2.119
2.730

0.000
0.000
0.017
0.003

*1-tail test

Maurizio Pisati

Spatial Data Analysis in Stata

51/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Local indices of spatial autocorrelation

A local index of spatial autocorrelation expresses, for

each region ri of a given study area A, the degree of

similarity between that region and its neighboring regions
with respect to a numeric variable Y (Pfeiffer et al. 2008)

Maurizio Pisati

Spatial Data Analysis in Stata

52/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Local indices of spatial autocorrelation

A local index of spatial autocorrelation expresses, for

each region ri of a given study area A, the degree of

similarity between that region and its neighboring regions
with respect to a numeric variable Y (Pfeiffer et al. 2008)
The local indices of spatial autocorrelation are derived

from the corresponding global indices and share their

fundamental properties

Maurizio Pisati

Spatial Data Analysis in Stata

52/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Local indices of spatial autocorrelation in Stata

Stata users can compute local indices of spatial auto-

correlation using spatlsa, a user-written command

published in the Stata Technical Bulletin (Pisati 2001)

Maurizio Pisati

Spatial Data Analysis in Stata

53/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Local indices of spatial autocorrelation in Stata

Stata users can compute local indices of spatial auto-

correlation using spatlsa, a user-written command

published in the Stata Technical Bulletin (Pisati 2001)
spatlsa (latest version: 1.0) computes four indices of

spatial autocorrelation: Morans Ii , Getis and Ords Gi and

G?i , and Gearys ci . For each index and each region in the
analysis, spatlsa computes and displays in tabular form
the value of the index itself, the expected value of the index
under the null hypothesis of no local spatial autocorrelation, the standard deviation of the index, the
z -value, and the corresponding one- or two-tailed p-value

Maurizio Pisati

Spatial Data Analysis in Stata

53/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Local indices of spatial autocorrelation: example

Study area: Ohio

Regions: 88 counties
Variable of interest: Pct. population aged 18+ with

poor-to-fair health status (pct poorhealth)

Maurizio Pisati

Spatial Data Analysis in Stata

54/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Local indices of spatial autocorrelation: example

We use spatlsa with the standardized spatial weights matrix Ws

previously generated by spatwmat to compute Morans Ii on the
variable of interest. In the output, counties are sorted by z -value

Maurizio Pisati

Spatial Data Analysis in Stata

55/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Local indices of spatial autocorrelation: example

We use spatlsa with the standardized spatial weights matrix Ws

previously generated by spatwmat to compute Morans Ii on the
variable of interest. In the output, counties are sorted by z -value
!"#$%&#$'(!)*+',-.(*$)/!0-.*$)+()$123$#,!'*#&/"4!#'!$#*3#53)6/'
(!/',-.(*$)/!07#$#23$#,!'89/#5'
!"#$9!#'"8$:"..5;/#9$;!'%"4!#'&.5#*')3"*#&/#'!.5$!

Maurizio Pisati

Spatial Data Analysis in Stata

55/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Local indices of spatial autocorrelation: example

User: Maurizio

Measures of local spatial autocorrelation

(Output omitted)
Moran's Ii (Poor-to-fair health status (pct. pop 18+))
name
Knox
Hardin
Paulding
Licking

E(Ii)

sd(Ii)

p-value*

-0.816
-0.760
-1.089
-0.457

Ii

-0.011
-0.011
-0.011
-0.011

0.358
0.358
0.560
0.358

-2.246
-2.090
-1.924
-1.244

0.012
0.018
0.027
0.107

0.545
0.927
0.677
0.908
0.949
1.087
1.246
2.433
2.197
3.578
5.503
3.911
5.400

-0.011
-0.011
-0.011
-0.011
-0.011
-0.011
-0.011
-0.011
-0.011
-0.011
-0.011
-0.011
-0.011

0.358
0.560
0.389
0.482
0.358
0.358
0.389
0.482
0.429
0.482
0.560
0.389
0.482

1.555
1.675
1.769
1.906
2.682
3.067
3.230
5.069
5.150
7.442
9.844
10.077
11.220

0.060
0.047
0.038
0.028
0.004
0.001
0.001
0.000
0.000
0.000
0.000
0.000
0.000

(Output omitted)
Hancock
Williams
Delaware
Mercer
Putnam
Henry
Vinton
Gallia
Pike
Adams
Lawrence
Jackson
Scioto
*1-tail test

Maurizio Pisati

Spatial Data Analysis in Stata

56/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Overview
Measuring spatial autocorrelation
Global indices of spatial autocorrelation
Local indices of spatial autocorrelation

Local indices of spatial autocorrelation: example

Williams
Henry
Paulding
Putnam

Hardin
Mercer
Knox
Delaware

Vinton
Pike
Adams

Scioto

Jackson
Gallia
Lawrence

Maurizio Pisati

Spatial Data Analysis in Stata

57/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Fitting spatial regression models

Maurizio Pisati

Spatial Data Analysis in Stata

58/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial regression

The aim of spatial regression is to estimate the

relationship between an outcome variable of interest Y and

one or more predictors X, taking into proper account the
spatial dependence among observations

Maurizio Pisati

Spatial Data Analysis in Stata

59/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial regression

The aim of spatial regression is to estimate the

relationship between an outcome variable of interest Y and

one or more predictors X, taking into proper account the
spatial dependence among observations
Two types of spatial dependence are most commonly
considered (Ward and Gleditsch 2008):
A spatial autoregressive process in the error term
A spatial autoregressive process in the outcome variable

Maurizio Pisati

Spatial Data Analysis in Stata

59/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial error model

The first type of spatial dependence is represented by the

spatial error model:

Y = X + W + 
where Y denotes an N 1 vector of observations on the
outcome variable; X denotes an N j matrix of
observations on the predictor variables; denotes a j 1
vector of regression coefficients; denotes the spatial
autoregressive parameter; W denotes the N N spatial
weights matrix; denotes an N 1 vector of spatial errors;
and  denotes an N 1 vector of normally distributed,
homoskedastic, and uncorrelated errors
Maurizio Pisati

Spatial Data Analysis in Stata

60/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial lag model

The second type of spatial dependence is represented by

the spatial lag model:

Y = X + WY + 
where denotes the spatial autoregressive parameter; and
all the other terms are defined as above

Maurizio Pisati

Spatial Data Analysis in Stata

61/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial error vs. spatial lag models

The spatial lag model treats spatial dependence as

substance, assuming that the value taken by Y in each

region is affected by the values taken by Y in the
neighboring regions

Maurizio Pisati

Spatial Data Analysis in Stata

62/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial error vs. spatial lag models

The spatial lag model treats spatial dependence as

substance, assuming that the value taken by Y in each

region is affected by the values taken by Y in the
neighboring regions
On the other hand, the spatial error model treats spatial

dependence as nuisance

Maurizio Pisati

Spatial Data Analysis in Stata

62/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial regression in Stata

Stata users can fit spatial error and spatial lag models

using spatreg, a user-written command published in the

Stata Technical Bulletin (Pisati 2001)

Maurizio Pisati

Spatial Data Analysis in Stata

63/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial regression in Stata

Stata users can fit spatial error and spatial lag models

using spatreg, a user-written command published in the

Stata Technical Bulletin (Pisati 2001)
An excellent alternative to spatreg is represented by

sppack, a suite of Stata commands freely available from

the SSC Archive written by David M. Drukker, Hua
Peng, Ingmar Prucha, and Rafal Raciborski

Maurizio Pisati

Spatial Data Analysis in Stata

63/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

Spatial regression in Stata

Stata users can fit spatial error and spatial lag models

using spatreg, a user-written command published in the

Stata Technical Bulletin (Pisati 2001)
An excellent alternative to spatreg is represented by

sppack, a suite of Stata commands freely available from

the SSC Archive written by David M. Drukker, Hua
Peng, Ingmar Prucha, and Rafal Raciborski
sppack is faster and more flexible than spatreg. Moreover,

while spatreg is limited to the analysis of small sets of

observations, sppack can deal with very large N s

Maurizio Pisati

Spatial Data Analysis in Stata

63/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

References

Maurizio Pisati

Spatial Data Analysis in Stata

64/65

Introduction
Visualizing spatial data
Exploring spatial point patterns
Measuring spatial proximity
Detecting spatial autocorrelation
Fitting spatial regression models

References
Bailey, T.C. and A.C. Gatrell. 1995. Interactive Spatial Data Analysis.
Harlow: Longman.

Pfeiffer, D., Robinson, T., Stevenson, M., Stevens, K., Rogers, D. and A.
Clements. 2008. Spatial Analysis in Epidemiology. Oxford: Oxford
University Press.

Pisati, M. 2001. sg162: Tools for spatial data analysis. Stata Technical
Bulletin 60: 2137. In Stata Technical Bulletin Reprints, vol. 10, 277298.
College Station, TX: Stata Press.

Slocum, T.A., McMaster, R.B., Kessler, F.C. and H.H. Howard. 2005.
Thematic Cartography and Geographic Visualization. 2nd ed. Upper Saddle
River, NJ: Pearson Prentice Hall.

Tobler, W.R. 1970. A computer movie simulating urban growth in the

Detroit region. Economic Geography 46: 234240.

Waller, L.A. and C.A. Gotway. 2004. Applied Spatial Statistics for Public
Health Data. Hoboken, NJ: Wiley.

Ward, M.D. and K.S. Gleditsch. 2008. Spatial Regression Models. Thousand
Oaks, CA: Sage.
Maurizio Pisati

Spatial Data Analysis in Stata

65/65