iterCIR {cir}

Centered-isotonic-regression (CIR): iterative version for better bias correction

Description

EXPERIMENTAL: Nonparametric forward point estimation of a monotone response (y) as a function of dose (x), using an iterative version of the centered-isotonic-regression (CIR) algorithm. The code works, but delivers marginal improvement at greater computational cost (an issue if you simulate a large ensemble), and somewhat convoluted interpretation. Use at your own risk. For explanation, see Oron and Flournoy (2017), Section 3.2.

Usage

iterCIR(y, outx = NULL, tol = 0.001, maxit = 10, full = FALSE, ...)

Arguments

y

See cirPAVA

outx

vector of x values for which predictions will be made. If NULL (default), this will be set to the set of unique values in the x argument (or equivalently in y$x).

tol

The iteration's convergence tolerance level (default 1e-3)

maxit

integer, maximum number of iterations (default 10)

full

logical, is a more complete output desired? if FALSE (default), only a vector of point estimates for y at the provided dose levels is returned

...

Other arguments passed on to cirPAVA

Value

under default, returns a vector of y estimates at unique x values. With full=TRUE, returns a list of 3 doseResponse objects named output,input,shrinkage for the output data at dose levels, the input data, and the function as fit at algorithm-generated points, respectively.

See Also

cirPAVA,quickIsotone