Message-ID: <500e85eb.a652b40a.4e88.fffff936@mx.google.com>
Date: 2012-07-24T11:24:18Z
From: Luigi Marongiu
Subject: limit of detection (LOD) by logistic regression
In-Reply-To: <A4E5A0B016B8CB41A485FC629B633CED4694ECB29D@GOLD.corp.lgc-group.com>
Dear Ellison,
You are right, now the figure is good! Question solved.
Thank you very much!
Best wishes,
Luigi
-----Original Message-----
From: S Ellison [mailto:S.Ellison at LGCGroup.com]
Sent: 24 July 2012 10:20
To: Luigi; r-help at r-project.org
Subject: RE: [R] limit of detection (LOD) by logistic regression
> set 1; however the figure obtained from the sample set 2
> shows that interpolation is not correct.
I don't think the interpolation is incorrect; what is making it look
incorrect is using a straight line to represent a logistic regression.
Try adding the predicted values for the line to your plot:
lines( dil <- 10^seq(-1, 6, 0.05), predict(model,
newdata=data.frame(dilution=dil), type="response"))
S Ellison
*******************************************************************
This email and any attachments are confidential. Any use...{{dropped:8}}