pom {paltran}R Documentation

Proportional Odds modelling for paleolimnology

Description

Proportional odds modells are used to model the distribution of diatoms along environmental gradients. Than these models are applied to samples of a sediment core. This function is just a test version! The algorithm is slow, one run with the given data takes about 5 to 10 minutes.

Usage


pom(..., d.plot = "TRUE")

Arguments

... x,y,z: required: species training set (x) as matrix and related environmental parameter (y). optional: test set(z) - species data from a sediment core
d.plot TRUE/FALSE: if TRUE diagnostic plots are given at the end of the analysis

Details

The relative abundances of a taxon within a data set is transformed to abundance classes (class 1 = <0 to <1 class 2 = 1 < 5 ( frequent), class 6 = 60 - 100 In contrast to WA, WA-LS and MW the POM function use the abundance class specific optimum of a species, i.e. when a taxa occurred in a sediment core within the class 1, it might be interesting, where the optimum for this specific species abundance class to a related environmental factor in the training set is. This optimum can than be used to infer past environmental parameters. In general: Not the overall species optimum is used to infer environmental parameter, but several different optima were estimate related to different abundance classes. The pom() function uses mainly the function polr from the package MASS (Venables, W.N. Ripley, B.D. (2002). Modern applied statistics with S-Plus (third edition). Springer, New York, 501 pp.). WARNING: This version was build under R 2.5.1 - a test run with 2.7.0 leads to warning massages using the mgcv package! Use mgcv 1.3-31 to avoid the warning massage.

Value

env.train environmental parameter of the train set
env.test.pred parameter on which the prediction is done
env.matrix.train density curve for the inferred environmental parameter for the train set
inf.env.train inferred environmental parameter for the train set
env.matrix.test density curve for the inferred environmental parameter for the test set
reconstruction inferred environmental parameter for the test set
inf.env.cross_train inferred environmental parameter for the train set using cross validation
performance performance of the model (R2,RMSE,RMSEP,.....

Author(s)

Sven Adler, sven.adler2@uni-rostock.de, University Rostock, Institute for Biosciences, General and Systematic Botany, Germany

References

Adler,S., Huebener T.,Anderson J.N., Lotter,A.F.,Modelling the distribution of diatoms along contrasting environmental gradients in Europe using GLM, WA and Proportional Odds Models, submitted in Journal of European Phycology

See Also

wa, wapls, mwtraf

Examples

data(dud.df)
data(age_dud)
data(train_set.MV)
data(train_env.MV)

fit1<-pom(train_set.MV,train_env.MV,dud.df)
fit1$performance
palplot(fit1$reconstruction)
palplot(fit1$reconstruction,age_dud)

[Package paltran version 1.0-0 Index]