Closed osorensen closed 5 years ago
Here are illustrations of print and coef methods implemented in development version 0.2.3.9001. This is shown for the case of gmus
, but similar S3 methods are implemented for the other algorithms (gmu_lasso
, corrected_lasso
, cv_corrected_lasso
).
library(hdme)
# Example with linear regression
set.seed(1)
n <- 100 # Number of samples
p <- 10 # Number of covariates
# True (latent) variables
X <- matrix(rnorm(n * p), nrow = n)
# Measurement matrix (this is the one we observe)
W <- X + matrix(rnorm(n*p, sd = 1), nrow = n, ncol = p)
# Coefficient vector
beta <- c(seq(from = 0.1, to = 1, length.out = 5), rep(0, p-5))
# Response
y <- X %*% beta + rnorm(n, sd = 1)
# Run the MU Selector on a delta grid
fit1 <- gmus(W, y, delta = seq(from = 0, to = 0.5, by = 0.1))
# print method
fit1
#> Generalized MU Selector with family gaussian, with 10 variables fitted with regularization parameters lambda = 0.09732412 and 6 delta values.
#> Use functions plot() and coef() for more information about the fitted values.
# coef method
coef(fit1)
#> Number of nonzero coefficient estimates
#> as a function of regularization parameters
#> (lambda, delta):
#> lambda delta nonzeros
#> 0.097 0.0 7
#> 0.097 0.1 4
#> 0.097 0.2 5
#> 0.097 0.3 3
#> 0.097 0.4 3
#> 0.097 0.5 3
# Run the MU Selector on a single delta
fit2 <- gmus(W, y, delta = 0.2)
# coef method by default shows only nonzero coefficient estimates
coef(fit2)
#> Non-zero coefficient estimates at
#> regularization parameters (lambda, delta) = (0.081, 0.2):
#> coefficient estimate
#> 1 -5.690423e-17
#> 2 3.440890e-02
#> 3 1.158849e-01
#> 4 2.003944e-01
#> 5 3.659696e-01
# can optionally show all coefficient estimates
coef(fit2, all = TRUE)
#> Coefficient estimates at regularization
#> parameters (lambda, delta) = (0.081, 0.2):
#> coefficient estimate
#> 1 -5.690423e-17
#> 2 3.440890e-02
#> 3 1.158849e-01
#> 4 2.003944e-01
#> 5 3.659696e-01
#> 6 0.000000e+00
#> 7 0.000000e+00
#> 8 0.000000e+00
#> 9 0.000000e+00
#> 10 0.000000e+00
Created on 2019-05-14 by the reprex package (v0.2.1)
Add print, summary, and coef methods to the S3 classes defined.