{smcl}
{* 13Sep2009}{...}
{cmd:help distmatch}
{hline}
{title:Title}
{p2colset 5 18 20 2}{...}
{p2col :{hi: distmatch} {hline 2}}distance matching based on latitudes and longitudes{p_end}
{marker s_Syntax}
{title:Syntax}
{p 4 4 6}
{cmdab:distmatch} [{varlist}], {opt id( )} {opt lat:itude( )} {opt lon:gitude( )} [{it:options}]
{marker s_Description}
{title:Description}
{p 4 4 6}
{cmd:distmatch} provides a fast and easy way to perform distance matching based on latitudes and
longitudes. It mathces location identifiers, attributes, or distances from one to another. It also
counts the number of neighboring locations within a given area, either in circles or bands of
concentric circles.
{p 4 4 6}
Outputs produced by {cmd:distmatch} are typically used in studies of geocoded data. These outputs
can also be produced by non-Stata softwares such as ArcGIS and ArcView.
{p 4 4 6}
{cmd:distmatch} implements {browse `"http://en.wikipedia.org/wiki/Haversine_formula"':haversine} formula for distance.
{p 4 4 6}
Distance matching with full ranking of nearby neighbors can be computationally intensive. Observations of
3,000 may take several minutes to complete. Observations of 2,000,000 will take about 1 day per nearest neighbor
for each observations.
{p 4 4 6}
The default setting is to calculate distances for each observation in miles.
{marker s_Options}
{title:Options}
{dlgtab:Main}
{p 4 12 6}{opt id(varname)} unique identifiers. They cannot be missing. {p_end}
{p 4 12 6}{opt lat:itude(varname)} decimal-degree latitude. {p_end}
{p 4 12 6}{opt lon:gitude(varname)} decimal-degree longitude. {p_end}
{p 4 12 6}{opt don:or(varname)} an indicator for donor. Neighbors will be found for them. The default is 1 for everyone.{p_end}
{p 4 12 6}{opt rec:ipient(varname)} an indicator for recipients. They are candidate to be someone's neighbor. The default is 1 for everyone.{p_end}
{p 4 12 6}{opt near:est(#)} the number of nearest neighbors to be found. {opt near:est(.)} will make it look
until it runs out of potential neighbors. The default is 5.{p_end}
{p 4 12 6}{opt min:imum(#)} the minimum distance. {p_end}
{p 4 12 6}{opt max:imum(#)} the maximum distance.{p_end}
{p 4 12 6}{opt count} count the number of neighbors within the area upto {opt near:est(#)}{p_end}
{p 4 12 6}{opt km} kilometers instead of miles.{p_end}
{p 4 12 6}{opt noi:sily} display the progress on the screen.{p_end}
{dlgtab:Nomenclature}
{p 4 12 6}{opt dist( )} name of variables to contain distnaces. The default is "dist" {p_end}
{p 4 12 6}{opt count( )} name of variable to contain the number of neighbors. The default is "count".{p_end}
{p 4 12 6}{opt pre:fix( )} prefix. The default is "_". {p_end}
{p 4 12 6}{opt suf:fix( )} non-index suffix before the indices. The default is none. {p_end}
{title:Hints}
{p 4 4 6}If you have more than one datasets, they need to be appeneded into one dataset. They should
have the same name for longigutes and latitutudes so that they will appear in the same columns.
{p 4 4 6}Use donor( ) option to indicate the potential pool of neighbors. See an example below.
{p 4 4 6}Use recipient( ) option to indicate locations for which neighbors will be found. See an example below.
{p 4 4 6}Use noisily option to see the progress on the screen.
{marker s_0}
{title:Examples}
{p 4 4 6}* suggestions on how to create unique identifier{p_end}
{p 4 8 6}gen id = _n{p_end}
{p 4 8 6}egen id = group(state county){p_end}
{p 4 4 6}* You should download the ancillary file containing population centroids of U.S. counties{p_end}
{p 4 8 6}{stata `"net describe distmatch, from(http://fmwww.bc.edu/RePEc/bocode/d)"'} {p_end}
{p 4 8 6}{stata use distmatch_county.dta, clear} {p_end}
{p 4 4 6}* noisily produces the id and the distance to 5 nearest neighbors by default {p_end}
{p 4 8 6}{stata distmatch, id(id) lat(lat) lon(lon) } {p_end}
{p 4 8 6}{stata browse}{p_end}
{p 4 4 6}* extracts values (attributes) associated with the nearest neighbors {p_end}
{p 4 8 6}{stata drop _*} {p_end}
{p 4 8 6}{stata distmatch name lat longi, id(id) lat(lat) lon(lon)} {p_end}
{p 4 8 6}{stata browse}{p_end}
{p 4 4 6}* draw potential neighbors from a pool of donors and match them to the recipients {p_end}
{p 4 8 6}{stata drop _*} {p_end}
{p 4 8 6}{stata gen donor =1 in 1/1500}{p_end}
{p 4 8 6}{stata gen recipient =1 in 1501/3232}{p_end}
{p 4 8 6}{stata distmatch, id(id) lat(lat) lon(lon) don(donor) rec(recipient) dist(one_to_the_other)}{p_end}
{p 4 8 6}{stata browse}{p_end}
{p 4 4 6}* searches in a band of concentric circle between 10 and 11 miles inclusive (for obs 1-500){p_end}
{p 4 8 6}{stata clear} {p_end}
{p 4 8 6}{stata set maxvar 12000} {p_end}
{p 4 8 6}{stata use distmatch_county.dta, clear} {p_end}
{p 4 8 6}{stata keep in 1/500} {p_end}
{p 4 8 6}{stata distmatch, id(id) lat(lat) lon(lon) min(10) max(11) nearest(.) count(conc_1_2)}{p_end}
{p 4 8 6}{stata browse}{p_end}
{title:Acknowledgement}
{p 4 4 6}Austin Nichols provided suggestions on how to improve the package.
{title:Remarks}
{p 4 12 6}The word "long" is reserved in Stata. Use "lon" or something else instead.{p_end}
{p 4 12 6}{opt min:imum( )} and {opt max:imum( )} are inclusive, i.e. >= and <=.{p_end}
{p 4 12 6}Conflicts between options should be reported with detailed examples to the address below.{p_end}
{title:Author}
{p 4 4 6}Roy Wada{p_end}
{p 4 4 6}roywada@hotmail.com{p_end}