|
R Code and Tutorials
R is a free,
open-source platform for statistical computing. It is powerful and
constantly growing. It can be frustrating to learn, but is well worth the
effort. See the R
site for downloads and documentation.
Lora Murphy and Charlie
Canham have assembled two packages of R code to facilitate likelihood
analyses.
The Likelihood
Package contains a set of functions for maximum likelihood estimation
using simulated annealing, a global optimization routine. The
implementation of simulated annealing is adapted from Goffe et al.
(1994), and allows bounded searches.
Windows binary: likelihood_1.4.zip
Unix/Linux binary: likelihood_1.4.tar.gz
For R >= 2.9
Right-click and choose "Save Target As"
(or choose save from the dialogue box that opens with a regular
click)
and download the zip file to your machine. NOTE: Do not unzip the file.
The NeighLikeli
Package contains functions for likelihood analyses of neighborhood
processes (i.e. Canham and Uriarte 2006), in which the function for the predicted
response variable contains terms that sum over the effect of a set of
neighbors (whose attributes and locations are contained within the data
frame).
Windows binary: neighlikeli_1.0.zip
Unix/Linux binary: neighlikeli_1.0.tar.gz
For R >= 2.9
Right-click and choose "Save Target As"
(or choose save from the dialogue box that opens with a regular
click)
and download the zip file to your machine. NOTE: Do not unzip the file.
Installing the
Packages:
Download the package zip file to a directory on your hard disk. Do not
UnZip the file. In R, use the "Install package(s) from local zip
file…" option under the Packages main menu to install the package
before first use. Then remember to "load" the package in any
work session when you want to use it.
For more details, see
the section on Package Installation in a brief
introduction to R we put together for the course.
References:
- Canham, C. D. and M.
Uriarte. 2006. Analysis of neighborhood dynamics of forest
ecosystems using likelihood methods and modeling. Ecological
Applications 16:62-73.
- Goffe, W. L., G. D.
Ferrier, and J. Rogers. 1994. Global optimization of statistical
functions with simulated annealing. Journal of Econometrics 60:
65-99.
|
