Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Nearmrg deleting unmerged observations
From
Austin Nichols <[email protected]>
To
[email protected]
Subject
Re: st: Nearmrg deleting unmerged observations
Date
Fri, 12 Jul 2013 17:04:51 -0400
Lara Loewenstein <[email protected]>:
All geocoding AFAIK is time-specific and very very recent (map data as
of last year at the earliest), but in places I am familiar with there
are not many eliminated or changed numbers, just additions, so a
geocode from 2012 maps should give the same answer as one from 2002
maps. Also, I was thinking -nearmrg- (SSC) used edit distance, but it
bases distance on a single numeric variable, which means it could not
do the matches the way I was proposing. Didn't the original program
use edit distances? Back in 2002 or so?
On Fri, Jul 12, 2013 at 4:19 PM, Lara Loewenstein <[email protected]> wrote:
> data is clean, so there shouldn't be many mismatches because of
> different street names. The problem is that there is almost a 20 year
> difference from when the using data and the master data were created.
> My concern that street numbers could have been eliminated or changed,
> and I wanted to look at the nearmrg results just to see if there were
> a lot of inexact, yet close matches. Do you know how time invariant or
> variant geocoding is?
>
> Eric: Thanks, that will work. I was hoping to avoid doing that, but it
> definitely solves the problem.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/