psi_Init: Initial value estimate for the psiLearn.

In mylzwq/psi-learning-for-ITR: An Implementation of psiLearn on Individualized Treatment Rule

Description

Initial value estimate for both psi-linear,psi-kernel,and psi-linear-VS.

Usage

psi_Init(X,A,R,tau=0.1,kappa=seq(0.01,1,length.out=10),sigma=NULL,kernel='linear',lambda=NULL,VS=FALSE)

Arguments

X	n by p input matrix.
A	a vector of n entries coded 1 and -1 for the treatment assignments.
R	a vector of outcome variable, larger is more desirable.
tau	tuning parameter for the loss function in psi-Learn
kappa	tunning parameter to control the complexity of the decision function, a seq of kappa is set as the default.
sigma	bandwidth parameter when the kernel is 'rbf',if not provided,can be estimated from Sig_est
kernel	kernel function for pai-Learn, can be 'linear' or 'rbf' (radial basis kernel), default is 'linear'. When 'rbf' is specified, one can specify the sigma parameter for radial basis kernel or estimate from Sig_est.
lambda	when using the variable selection method for inital estimate, lambda is regarded as the tunning parameter in controlling the complexity in L1 penalty
VS	wheter to use the variable selection method as the inital estimate,default is FALSE

Value

It returns initial estimated coefficients for psiLearn with the best tuning parameters picked by cross validation.

w	the coefficent for the decision function.
bias	the intercept in both the linear case and the kernel case.

Author(s)

MingyangLiu <[e-mail]>

See Also

tlp_S

mylzwq/psi-learning-for-ITR documentation built on May 15, 2019, 1:18 p.m.