Moran's I based on ZIP Code data

Check out

http://www2.census.gov/cgi-bin/shapefiles2009/national-files - left  
hand
side has national - level data.

All 5 digit zipcode files are based on 2002 data but zipcode  
boundaries
change less frequently than tracts and blockgroups, there may be  
some issues
in correctly mapping out areas in high growth regions of the U.S.  
with new
zipcodes.

Sharon

On Fri, Aug 20, 2010 at 12:47 PM, Michael Haenlein
<haenlein at escpeurope.eu>wrote:

Thanks very much for your reply, Roger!

I have downloaded the shape files from the US Census ZCTA webpage.  
In case
anyone else is interested in obtaining them the URL is:
http://www.census.gov/geo/www/cob/z52000.html#shp

I also managed to import those files into R and to convert them into
a neighbour list:

Alabama <-readShapePoly("c:/111/zt01_d00")
Alaska <-readShapePoly("c:/111/zt02_d00")
Arizona <-readShapePoly("c:/111/zt04_d00")
...

Alabama.nb <- poly2nb(Alabama)
Alaska.nb <- poly2nb(Alaska)
Arizona.nb  <- poly2nb(Arizona)
...

The problem is that instead of having one neighbour list I now have  
52 ones
(one for each state).
Is there a way to combine all of them into one large neighbour list  
which I
can then use as an input for my analysis?




-----Original Message-----
From: Roger Bivand [mailto:Roger.Bivand at nhh.no]
Sent: Thursday, August 19, 2010 23:54
To: Michael Haenlein
Cc: r-sig-geo at stat.math.ethz.ch
Subject: Re: [R-sig-Geo] Moran's I based on ZIP Code data

On Thu, 19 Aug 2010, Michael Haenlein wrote:

The first thing is to get the locations of the zip codes (about  
30,000?) -
they are published as shapefiles by state (US Census ZCTA), so a  
polygon
representation is possible, but you could also look for a point
representation. Next make a neighbour list (nb) object to the zip  
code
entities for which you have observations. Then you could use  
nb2blocknb()
in
spdep to "block up" observations where more than one belongs to the  
same
zip
code, which effectively makes all the observations in a zip code
neighbours,
and adds all the observations in neighbouring zip codes too.
It was written for housing data with only a postcode but no geocoded
address.

Hope this helps,

Roger


--
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian  
School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway.
voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no

      [[alternative HTML version deleted]]

	[[alternative HTML version deleted]]

Thanks very much for your reply, Roger!

I have downloaded the shape files from the US Census ZCTA webpage. In case
anyone else is interested in obtaining them the URL is:
http://www.census.gov/geo/www/cob/z52000.html#shp

I also managed to import those files into R and to convert them into
a neighbour list:

Alabama <-readShapePoly("c:/111/zt01_d00")
Alaska <-readShapePoly("c:/111/zt02_d00")
Arizona <-readShapePoly("c:/111/zt04_d00")
...

Alabama.nb <- poly2nb(Alabama)
Alaska.nb <- poly2nb(Alaska)
Arizona.nb  <- poly2nb(Arizona)
...

The problem is that instead of having one neighbour list I now have 52 ones
(one for each state).
Is there a way to combine all of them into one large neighbour list which I
can then use as an input for my analysis?

-----Original Message-----
From: Roger Bivand [mailto:Roger.Bivand at nhh.no]
Sent: Thursday, August 19, 2010 23:54
To: Michael Haenlein
Cc: r-sig-geo at stat.math.ethz.ch
Subject: Re: [R-sig-Geo] Moran's I based on ZIP Code data

On Thu, 19 Aug 2010, Michael Haenlein wrote:

The first thing is to get the locations of the zip codes (about 30,000?) -
they are published as shapefiles by state (US Census ZCTA), so a polygon
representation is possible, but you could also look for a point
representation. Next make a neighbour list (nb) object to the zip code
entities for which you have observations. Then you could use nb2blocknb() in
spdep to "block up" observations where more than one belongs to the same zip
code, which effectively makes all the observations in a zip code neighbours,
and adds all the observations in neighbouring zip codes too.
It was written for housing data with only a postcode but no geocoded
address.

Hope this helps,

Roger


--
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no

The national files for the zipcodes seems greyed out. I would caution 
against creating nb lists for each state separate and then creating a 
US wide neighbour list because, there will some zip codes in Alabama 
who are neighbours to  zipcodes in GA, MS, TN. I would merge them 
first into one big file and then construct the poly list.  you may run 
into memory issues for this operation, depending on your set up.


Nikhil Kaza
Asst. Professor,
City and Regional Planning
University of North Carolina

nikhil.list at gmail.com

On Aug 20, 2010, at 1:06 PM, Sharon O'Donnell wrote:

Check out

http://www2.census.gov/cgi-bin/shapefiles2009/national-files - left hand
side has national - level data.

All 5 digit zipcode files are based on 2002 data but zipcode boundaries
change less frequently than tracts and blockgroups, there may be some 
issues
in correctly mapping out areas in high growth regions of the U.S. 
with new
zipcodes.

Sharon

On Fri, Aug 20, 2010 at 12:47 PM, Michael Haenlein
<haenlein at escpeurope.eu>wrote:

Thanks very much for your reply, Roger!

I have downloaded the shape files from the US Census ZCTA webpage. 
In case
anyone else is interested in obtaining them the URL is:
http://www.census.gov/geo/www/cob/z52000.html#shp

I also managed to import those files into R and to convert them into
a neighbour list:

Alabama <-readShapePoly("c:/111/zt01_d00")
Alaska <-readShapePoly("c:/111/zt02_d00")
Arizona <-readShapePoly("c:/111/zt04_d00")
...

Alabama.nb <- poly2nb(Alabama)
Alaska.nb <- poly2nb(Alaska)
Arizona.nb  <- poly2nb(Arizona)
...

The problem is that instead of having one neighbour list I now have 
52 ones
(one for each state).
Is there a way to combine all of them into one large neighbour list 
which I
can then use as an input for my analysis?




-----Original Message-----
From: Roger Bivand [mailto:Roger.Bivand at nhh.no]
Sent: Thursday, August 19, 2010 23:54
To: Michael Haenlein
Cc: r-sig-geo at stat.math.ethz.ch
Subject: Re: [R-sig-Geo] Moran's I based on ZIP Code data

On Thu, 19 Aug 2010, Michael Haenlein wrote:

The first thing is to get the locations of the zip codes (about 
30,000?) -
they are published as shapefiles by state (US Census ZCTA), so a 
polygon
representation is possible, but you could also look for a point
representation. Next make a neighbour list (nb) object to the zip code
entities for which you have observations. Then you could use 
nb2blocknb()
in
spdep to "block up" observations where more than one belongs to the 
same
zip
code, which effectively makes all the observations in a zip code
neighbours,
and adds all the observations in neighbouring zip codes too.
It was written for housing data with only a postcode but no geocoded
address.

Hope this helps,

Roger


-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian 
School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway.
voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no

      [[alternative HTML version deleted]]

    [[alternative HTML version deleted]]