fcmml: Fisher score method for estimating the regression coeffcients...

Description Usage Arguments Value

Description

Fisher score method for estimating the regression coeffcients in a mixture of logisitc regression.

Usage

1
fcmml(Y, X, alpha, beta, T0S, T1S, converged = 1e-06, maxIT = 200)

Arguments

Y

The response variable. Must be binary.

X

The design matrix. Intercept included.

alpha

The initial values of the coefficients for the confounders.

beta

The initial value of the coefficient for the true somatic mutation value.

T0S

The weights vector for the mixture when the true somatic muation equals 0.

T1S

The weights vector for the mixture when the true somatic muation equals 1.

converged

The tolerance for the convergence. Default is 1e-6.

maxIT

The maximal number of the EM iteration times. Default is 200.

Value

The estimated regression coefficients.


Sun-lab/SAME documentation built on May 30, 2019, 11:48 p.m.