[R-sig-Geo] Prepping data for FlexScan

[Paul Lantos]

Hi,

 I'm trying to prepare data for FlexScan to use the flexible cluster scan statistic, and I'm having difficulty getting data in the right format.


 The problem seems to be in the creation of a neighbor matrix file. I've been using a point data source, creating thiessen polygons (in ArcGIS), then exporting as a shapefile to R where in spdep I create an nb object using poly2nb. I've then exported it using the nb2inla function.


 But FlexScan does not seem to like the neighbor matrix file I've created, it gives me a generic error.


 Has anyone successfully prepared data for FlexScan? Is there an alternative flexible scanning statistic? I am unaware of an R package that performs flexible cluster analysis, and I am not interested in circular or elliptical clusters (which I can do easily enough in SatScan).


Thanks,

Paul


______________________________________
Paul M. Lantos, M.D., FIDSA, FAAP, FACP
    Pediatric Infectious Diseases
    Hospital Medicine
Depts of Internal Medicine and Pediatrics
Duke University Medical Center

______________________________________

This message and any included attachments are confidential and are intended only for the addressee(s).  The information contained herein may be confidential under the attorney/client privilege and/or the quality assurance and peer review privilege.  Unauthorized review, forwarding, printing, copying or otherwise distributing or using such information is prohibited and may be unlawful.  If you have received this message in error, please notify the sender promptly by email or telephone and delete the message.

	[[alternative HTML version deleted]]