3

i am working on a consulting problem that involves error-in-variables (Deming) regression to compare two measurement techniques. The details are given in a question I asked on CV. Here is the link. Bill Huber pointed me to the mcr package on cran that will solve the Deming regression problem. I was asked to make a bootstrap correction because some of the data points could be correlated.

Can any expert in R on this site show me how I could imbed the Deming regression in a bootstrap procedure?

Community
  • 1
  • 1
Michael R. Chernick
  • 151
  • 8
  • @smilig. Thanks for editing my question and adding the name of the mrc package. – Michael R. Chernick Sep 29 '12 at 13:34
  • 3
    No problem @Michael. As a side note, people are usually more willing and interested to answer questions that provide some example data. Here's a reference point, if you haven't seen it yet: http://stackoverflow.com/a/5963610/1088647 – smillig Sep 29 '12 at 13:56
  • 2
    This is a link to a posting to Rhelp by Terry Thereau. It has the code for what he was calling generalized Deming regression based on an article by Ripley.http://markmail.org/message/4mo62jqfyudrchzi?q=list:org%2Er-project%2Er-help+deming+therneau . Despite his statement that the code would not remain attached ... it was. – IRTFM Sep 29 '12 at 18:39

0 Answers0