[R-sig-ME] Unevely-spaced time dependency in a multiple-river dataset

Rich Shepard rshepard at appl-ecosys.com
Wed Jan 3 17:31:13 CET 2018

On Wed, 3 Jan 2018, Alessandro Manfrin wrote:

> ->The main aim of the study is to analyse changes in the fish diversity
> index ratio (as delta (Unrestored-Restored)/Unrestored) over several years
> (that I grouped in categories to deal with data linearity (a=0-2 years
> after restoration; b =3-5 years after; c= and so on up to 20 years)


   With the vast amount of information available to you perhaps a Bayesian
approach would best fit your objectives. You might look at Pulkkinen, H.
2015. Embracing uncertainty in fisheries stock assessment using Bayesian
hierarchical models. Thesis at the University of Helsinki. It's available on
the Web.

   There are Bayesian time series models available, but if you want to stay
with glm there is the tscount package in CRAN: "tscount: An R Package for
Analysis of Count Time Series Following Generalized Linear Models" since
fish survey data often are in counts/proportions. Looks like your data are
in counts and not presence/absence.



More information about the R-sig-mixed-models mailing list