[R] Help with predict.lm

[(Ted Harding)]

On 20-Apr-05 Mike White wrote:
> Ted
> Thank you for giving me so much of your time to provide
> an explanation of the likelihood ratio approach to the
> calibration problem.  It has certainly provided me with a
> new insight into this problem. I will check out the
> references you mentioned to get a better understanding
> of the statistics.

Hi Mike.

You should also look at

Philip J. Brown & Rolf Sundberg
Confidence and Conflict in Multivariate Calibration
JRSS B, vol. 49 (1987), pp. 46-57

which greatly extends the profile likelihood/likelihood
based confidence interval approach, relates it also to
to Bayesian approaches, and contrasts both with other
methods.

Best wishes,
Ted.


--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 20-Apr-05                                       Time: 12:53:29
------------------------------ XFMail ------------------------------