cv.lm.main: A function for the number of binary rules in the main effect...
In SubgrpID: Patient Subgroup Identification for Clinical Drug Development

A function for the number of binary rules in the main effect AIM with continuous outcome

1 2	cv.lm.main(x, y, K.cv = 5, num.replicate = 1, nsteps, mincut = 0.1, backfit = F, maxnumcut = 1, dirp = 0)

`x`	the predictor matrix.
`y`	the vector of the continuous response variable.
`K.cv`	number of folds for CV.
`num.replicate`	number of CV iterations.
`nsteps`	the maximum number of binary rules to be included in the index.
`mincut`	the minimum cutting proportion for the binary rule at either end. It typically is between 0 and 0.2. It is the parameter in the functions of AIM package.
`backfit`	a logical argument indicating whether the existing cutpoints are adjusted after including new binary rule.
`maxnumcut`	the maximum number of binary splits per predictor.
`dirp`	a vector for pre-specified direction of the binary split for each of the predictors. 0 represents "no pre-given direction"; 1 represents "(x>cut)"; -1 represents "(x<cut)". Alternatively, "dirp=0" represents that there is no pre-given direction for any of the predictor.

returns optimal number of binary rules based on CV along with CV score test statistics for the main effect, pre-validated score test statistics and prevalidated fits for individual observation.

SubgrpID documentation built on May 2, 2019, 8:02 a.m.