linear.nonmyop: Allocate treatments according to an information matrix based...

In mst1g15/biasedcoin:

Description

Allocate treatments according to an information matrix based optimality criterion allowing for a non-myopic approach. We assume a linear model for the response and simulate responses sequentially.

Usage

linear.nonmyop(covar, init, z.probs, k = NULL, N, int = NULL,
  lossfunc = calc.D, stoc, ...)

Arguments

covar	a dataframe for the covariates
init	the number of units in the initial design
z.probs	probability of each covariate being equal to 1
k	integer for number of "outer" loops in coordinate exchange algorithm for initial design
N	natural number greater than 0 for horizon
int	set to T if you allow for treatment-covariate interactions in the model, NULL otherwise
lossfunc	the objective function to minimize
stoc	set to T if treatments are allocated using a stochastic method where the probability is determined by the optimality crtierion. Set to F if treatments are allocated deterministically.
...	further arguments to be passed to <lossfunc>

Value

Design matrix D, all estimates of beta, final estimate of beta, responses y

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