[R-sig-ME] Robust SEs in GLMMs
chris at trickysolutions.com.au
Thu Nov 27 03:18:18 CET 2014
I faced the same problems a few years ago and tried using one of R's GAM
packages to calculate robust SE's. At the time we couldn’t get it to work
and I believe it was because we had too much data, so R ran into memory
Some other references you might like to have a look at are below. I
haven’t looked at them in detail myself yet, but as I've got a big data
set I intend to do this type of analysis on one day I've got a bit of "To
Read" list going
Pang (2010) Modeling Heterogeneity and Serial Correlation in Binary
Timeseries Crosssectional data. A Bayesian multilevel model with ARp
Jay Ver Hoef's package glmmLDTS . I think it deals with all your issues
(binomial, random effects, temporal autocorrelation...but not GAMM). It
may only run on < R.3.0):
*Ver Hoef, J.M.*, London, J.M., and Boveng, P.L. 2010. Fast computing of
some generalized linear mixed pseudo-models with temporal autocorrelation.
*Computational Statistics **25*(1)*: *39 - 55*. * DOI:
London, J.M., *Ver Hoef, J.M.*, Jeffries, S.J., Lance, M.M., and Boveng,
P.L. 2012. Haul-out behavior of harbor seals (Phoca vitulina) in Hood
Canal, Washington. *PLoS ONE **7*(6):e38180.
Chris Howden B.Sc. (Hons) GStat.
Data Analysis, Modelling and Training
Evidence Based Strategy/Policy Development, IP Commercialisation and
(mobile) +61 (0) 410 689 945
chris at trickysolutions.com.au
Disclaimer: The information in this email and any attachments to it are
confidential and may contain legally privileged information. If you are
not the named or intended recipient, please delete this communication and
contact us immediately. Please note you are not authorised to copy, use or
disclose this communication or any attachments without our consent.
Although this email has been checked by anti-virus software, there is a
risk that email messages may be corrupted or infected by viruses or other
interferences. No responsibility is accepted for such interference. Unless
expressly stated, the views of the writer are not those of the company.
Tricky Solutions always does our best to provide accurate forecasts and
analyses based on the data supplied, however it is possible that some
important predictors were not included in the data sent to us. Information
provided by us should not be solely relied upon when making decisions and
clients should use their own judgement.
From: r-sig-mixed-models-bounces at r-project.org
[mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf Of Sharon
Sent: Monday, 24 November 2014 8:05 AM
To: r-sig-mixed-models at r-project.org
Subject: [R-sig-ME] Robust SEs in GLMMs
When computing resource selection functions for animal telemetry data with
a binary response variable, where the 1s represent animal location data,
which are spatially and temporally correlated, and the 0s represent random
locations, which are not correlated, it is recommended to calculate
robust, or empirical, standard errors instead of using the model-based
standard errors to account for this differing correlation structure. As
far as I can tell, none of the glmm packages in R calculate these robust
SEs. Does anyone know of a way to use glmms that calculate these?
[[alternative HTML version deleted]]
R-sig-mixed-models at r-project.org mailing list
More information about the R-sig-mixed-models