future.loss: Assuming the currrent response, future covariate value and...

In mst1g15/biasedcoin:

Description

Assuming the currrent response, future covariate value and future treatment, Calculate optimality if horizon is 1. If not, iterate back to exp.loss function.

Usage

future.loss(z.next, t.next, zp, N, design, int, lossfunc, beta, y, bayes,
  dyn = NULL, ...)

Arguments

z.next	vector of covariate values for future unit
t.next	treatment of future unit
zp	vector of probabilities for each level of covariate z (needs to in the same order as all.z below)
N	natural number greater than 0 for horizon
design	design matrix constructed for all units up until the current unit
int	set to NULL if there are no interactions, set to T of there are interactions
lossfunc	the objective function to minimize
beta	estimate of the regression coefficients
y	responses that have been observed up until the current unit
bayes	set to T if bayesglm is used instead of glm. Default prior assumed.
dyn	set to T if there is a dynamic covariate
...	further arguments to be passed to <lossfunc>

Value

value of objective function assuming current response, future covariate value and future treatment

mst1g15/biasedcoin documentation built on Nov. 26, 2019, 4:01 a.m.