Compositional (version 1.0)

esov.compreg: ESOV regression

Description

Regression based on the ESOV divergence.

Usage

esov.compreg(y, x, B = 1000, ncores = 1, xnew = NULL)

Arguments

y

A matrix with the compositional data (dependent variable). Zero values are allowed.

x

The predictor variable(s), they have to be continuous.

B

If B is greater than 1 bootstrap estimates of the standard error are returned. If B=1, no standard errors are returned.

ncores

If ncores is 2 or more parallel computing is performed. This is to be used for the case of bootstrap. If B=1, this is not taken into consideration.

xnew

If you have new data use it, otherwise leave it NULL.

Value

• A list including:
• beta
  The beta coefficients.
• seb
  The standard error of the beta coefficients, if bootstrap is chosen, i.e. if B > 1.
• est
  The fitted or the predicted values (if xnew is not NULL).

Details

The ESOV metric is adopted as the objective function. This involves numerical optimisation. There is no log-likelihood.

References

Michail Tsagris (2015). A novel, divergence based, regression for compositional data. Proceedings of the 28th Panhellenic Statistics Conference. http://arxiv.org/pdf/1511.07600v1.pdf Endres, D. M. and Schindelin, J. E. (2003). A new metric for probability distributions. Information Theory, IEEE Transactions on, 49(7):1858-1860. Osterreicher, F. and Vajda, I. (2003). A new class of metric divergences on probability spaces and its applicability in statistics. Annals of the Institute of Statistical Mathematics, 55(3):639-653.

See Also

diri.reg, kl.compreg, ols.compreg, comp.reg, alfa.reg

Examples

library(MASS)
x <- fgl[, 1]
y <- fgl[, 2:9]
mod1 <- esov.compreg(y, x, B = 1, ncores = 1)
mod2 <- kl.compreg(y, x, B = 1, ncores = 1)
mod1
mod2