Re: Standard deviations and Confidence intervals with respect to
- To: mathgroup at smc.vnet.net
- Subject: [mg102678] Re: Standard deviations and Confidence intervals with respect to
- From: Ray Koopman <koopman at sfu.ca>
- Date: Tue, 18 Aug 2009 06:09:53 -0400 (EDT)
- References: <h3n5sr$2ed$1@smc.vnet.net> <h3pes0$oha$1@smc.vnet.net>
On Aug 17, 5:06 am, Alexey <lehi... at gmail.com> wrote:
> Thank you very much! This method is really that I have searched
> for. I would like to use it in my dissertation. How should I cite
> this method and your implementation of this method for Mathematica?
> What a publication should I cite?
I'm sorry, but I don't have any references for you. The model is
standard. Check in Cheng & Van Ness, Statistical Regression with
Measurement Error, Wiley, 1999, ISBN 978-0-340-61461-7.
The weighted least squares estimates can also be characterized
as maximum likelihood or minimum Mahalanobis distance estimates.
The estimates are obtained by alternating least squares:
estimate {a,b} given z, estimate z given {a,b}, ... .
The jackknifed estimate of the regression parameters, and the
corresponding estimate of their covariance matrix, are also standard.
Check with your department regarding the form in which material from
the internet should be cited. (My department follows APA format, which
is very specific about such things, and our university library checks
all theses meticulously to make sure they follow all the picky rules.)