Constrained linear least squares for compositional responses and predictors
Constrained linear least squares for compositional responses and predictors.
ols.compcomp(y, x, rs = 5, tol = 1e-4, xnew = NULL)
y |
A matrix with the compositional data (dependent variable). Zero values are allowed. |
x |
A matrix with the compositional predictors. Zero values are allowed. |
rs |
The number of times to run the constrained optimisation using different random starting values each time. |
tol |
The threshold upon which to stop the iterations of the constrained optimisation. |
xnew |
If you have new data use it, otherwise leave it NULL. |
The function performs least squares regression where the beta coefficients are constained to be positive and sum to 1. We were inspired by the transformation-free linear regression for compositional responses and predictors of Fiksel, Zeger and Datta (2020).
A list including:
runtime |
The time required by the regression. |
mse |
The mean squared errors. |
be |
The beta coefficients. |
est |
The fitted of xnew if xnew is not NULL. |
Michail Tsagris.
R implementation and documentation: Michail Tsagris mtsagris@uoc.gr.
Jacob Fiksel, Scott Zeger and Abhirup Datta (2020). A transformation-free linear regression for compositional outcomes and predictors. https://arxiv.org/pdf/2004.07881.pdf
library(MASS) set.seed(1234) y <- rdiri(214, runif(4, 1, 3)) x <- as.matrix(fgl[, 2:9]) x <- x / rowSums(x) mod <- ols.compcomp(y, x, rs = 1) mod
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.