logL

Computes the log-likelihood of the fitted regression parameters given the data
observed. Intended for internal use only.

Fit multinomial logistic regression with a penalty on the nuclear
norm of the estimated regression coefficient matrix, using proximal
gradient descent.

Scott Powers

npmr

Nuclear Penalized Multinomial Regression

logL function

<dl> <dt>B</dt>
<dd>Regression coefficient matrix</dd> <dt>b</dt>
<dd>Regression intercept vector</dd> <dt>X</dt>
<dd>Covariate matrix</dd> <dt>Y</dt>
<dd>Multinomial response matrix</dd></dl>

Arguments

Scott Powers, Trevor Hastie, Rob Tibshirani

Author

Log-likelihood for multinomial regression model — logL

<dl>

 <dt>B</dt>
<dd>Regression coefficient matrix</dd>

 <dt>b</dt>
<dd>Regression intercept vector</dd>

 <dt>X</dt>
<dd>Covariate matrix</dd>

 <dt>Y</dt>
<dd>Multinomial response matrix</dd>

</dl>

logL: Log-likelihood for multinomial regression model

Description

Usage

Value

Arguments

Author

See Also