discriminant_beta: Discriminant function

In gmmsslm: Semi-Supervised Gaussian Mixture Model with a Missing-Data Mechanism

Discriminant function

Description

Discriminant function in the particular case of g=2 classes with an equal-covariance matrix

Usage

discriminant_beta(pi, mu, sigma)

Arguments

pi	A g-dimensional vector for the initial values of the mixing proportions.
mu	A p \times g matrix for the initial values of the location parameters.
sigma	A p\times p covariance matrix.

Details

Discriminant function in the particular case of g=2 classes with an equal-covariance matrix can be expressed

d(y_i,\beta)=\beta_0+\beta_1 y_i,

where \beta_0=\log\frac{\pi_1}{\pi_2}-\frac{1}{2}\frac{\mu_1^2-\mu_2^2}{\sigma^2} and \beta_1=\frac{\mu_1-\mu_2}{\sigma^2}.

Value

beta0	An intercept of discriminant function
beta	A coefficient of discriminant function

gmmsslm documentation built on June 8, 2025, 2 p.m.