Date:         Thu, 17 Jul 1997 10:45:51 +1000
Reply-To:     Tim CHURCHES <TCHUR@DOH.HEALTH.NSW.GOV.AU>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Tim CHURCHES <TCHUR@DOH.HEALTH.NSW.GOV.AU>
Subject:      LINKPro System -Reply
Comments: To: tlin@MAIL.DOH.STATE.AR.US
Content-Type: text/plain

Tsaimei Lin asks:
>>> Tsaimei Lin <tlin@MAIL.DOH.STATE.AR.US> 17/July/1997 07:02am
>>>
I learned that LINKPro is an integrated SAS application system
for probabilistic and deterministic record linkage.
 
Have anybody used LINKPro System? What are the pros and cons?
What's peformance results for running it? Where can it be purchased?
How much will it cost?  What kind performance feature they have?
 
LINKPro run on UNIX workstation or MVS mainframe, can it be run
under WINDOWS NT 4.0?
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
I think that LINKPro is a probably a development of a set of SAS macros
developed by Wadja and Roos - see the Medline record below for the
full reference and abstract. We looked at the original Wadja and Roos
LINKS macros and found that although they worked OK, they were
slow and did not implement the full Fellegi and Sunter probabilistic record
linkage framework. If anyone has more information, please post to the
list.
 
We do quite a lot of probabilistic record linkage using two packages
from MatchWare Technologies (see http://www.matchware.com):
AutoStan and AutoMatch. These work with plain ASCI files which we
export from SAS datasets and then read back in to SAS datasets. The
MatchWare software implements the full Fellegi and Sunter model with
numerous refinements (both practical and theoretical) and runs very
fast.
 
Tim Churches
Epidemiology and Health Surveillance Branch
NSW Health Department
Sydney, Australia
Email: tchur@doh.health.nsw.gov.au
 
Medline record:
 
TITLE
     Simplifying record linkage: software and strategy.
AUTHOR(S)
     Wajda-A; Roos-LL
SOURCE (BIBLIOGRAPHIC CITATION)
     Comput-Biol-Med.1987; 17(4): 239-48.
INTERNATIONAL STANDARD SERIAL NUMBER
     0010-4825
LANGUAGE OF ARTICLE
     ENGLISH
ABSTRACT
     Although the methodology of record linkage is fairly well developed,
there is a need for less expensive methods and simpler software to
facilitate trying
     out different tactics to generate good linkages. The present work has
built on a fourth generation language SAS (Statistical Analysis System)
with
     accompanying macroprocessor, to develop a user-friendly and
flexible system for both exact and probabilistic matching. The major
features of the
     LINKS system are presented and illustrated using 1979-1984
information from the Manitoba Health Services Commission (MHSC)
registry file with the
     Canadian Mortality Data Base. Initial runs with exact, then
probabilistic, matching linked approximately 91% of the Vital Statistics
records to
     corresponding MHSC records. Subsequent modification of
parameters improved the linkage to 95%.
MEDLINE ACCESSION NUMBER
     88028544