glmnetcr-package: Penalized Constrained Continuation Ratio Models for Ordinal...

In glmnetcr: Fit a Penalized Constrained Continuation Ratio Model for Predicting an Ordinal Response

Description

This package provides a function glmnetcr for fitting penalized constrained continuation ratio models for predicting an ordinal response and associated methods for plotting, printing, extracting predicted classes and probabilities, and extracting estimated coefficients for selected models in the regularization path.

Details

The DESCRIPTION file: This package was not yet installed at build time.

Index: This package was not yet installed at build time.
This package contains functions for fitting penalized constrained continuation ratio models and extracting estimated coefficients, predicted class, and fitted probabilities. The model and methods can be used when the response to be predicted is ordinal, and is particularly relevant when there are more covariates than observations.

Author(s)

Kellie J. Archer Kellie J. Archer <[e-mail]>

Maintainer: Kellie J. Archer <[e-mail]> Kellie J. Archer <[e-mail]>

References

Archer K.J., Williams A.A.A. (2012) L1 penalized continuation ratio models for ordinal response prediction using high-dimensional datasets. Statistics in Medicine, 31(14), 1464-74.

See Also

See also glmnet ~~

Examples

data(diabetes)
x <- diabetes[, 2:dim(diabetes)[2]]
y <- diabetes$y
glmnet.fit <- glmnetcr(x, y)
AIC <- select.glmnetcr(glmnet.fit, which="AIC")
fitted(glmnet.fit, s=AIC)

Example output

Loading required package: glmnet
Loading required package: Matrix
Loading required package: foreach
Loaded glmnet 2.0-16

Warning message:
from glmnet Fortran code (error code -26); Convergence for 26th lambda value not reached after maxit=100 iterations; solutions for larger lambdas returned 
$BIC
     s23 
33.76674 

$AIC
     s23 
29.05452 

$class
 [1] "control"                  "control"                 
 [3] "control"                  "control"                 
 [5] "control"                  "control"                 
 [7] "control"                  "control"                 
 [9] "impaired fasting glucose" "control"                 
[11] "impaired fasting glucose" "impaired fasting glucose"
[13] "impaired fasting glucose" "impaired fasting glucose"
[15] "impaired fasting glucose" "type 2 diabetes"         
[17] "type 2 diabetes"          "type 2 diabetes"         
[19] "type 2 diabetes"          "type 2 diabetes"         
[21] "type 2 diabetes"          "type 2 diabetes"         
[23] "type 2 diabetes"          "type 2 diabetes"         

$probs
          control impaired fasting glucose type 2 diabetes
 [1,] 0.637759660                0.2968232      0.06541710
 [2,] 0.742789960                0.2154169      0.04179319
 [3,] 0.684967374                0.2608598      0.05417284
 [4,] 0.823974148                0.1494688      0.02655706
 [5,] 0.678639532                0.2657438      0.05561666
 [6,] 0.724791677                0.2297116      0.04549669
 [7,] 0.711520081                0.2401679      0.04831200
 [8,] 0.721282138                0.2324838      0.04623405
 [9,] 0.184716536                0.5184572      0.29682629
[10,] 0.484526771                0.4040373      0.11143589
[11,] 0.133808976                0.5043899      0.36180108
[12,] 0.162057619                0.5146922      0.32325022
[13,] 0.106676146                0.4864906      0.40683329
[14,] 0.188845436                0.5187891      0.29236544
[15,] 0.208919007                0.5190637      0.27201725
[16,] 0.022632656                0.3065999      0.67076742
[17,] 0.024619264                0.3163697      0.65901106
[18,] 0.010433956                0.2251335      0.76443251
[19,] 0.008090478                0.2021177      0.78979178
[20,] 0.024136568                0.3140570      0.66180641
[21,] 0.007500334                0.1956351      0.79686456
[22,] 0.024556421                0.3160707      0.65937286
[23,] 0.012003293                0.2386296      0.74936710
[24,] 0.007139356                0.1915041      0.80135657

glmnetcr documentation built on July 8, 2020, 6:21 p.m.