IBoost: Regression parameter estimation using I-Boost

In alexwky/I-Boost: An Integrative Boosting Approach for Predicting Survival Time With Multiple Genomics Platforms

Description

Takes in data matrices and performs I-Boost-CV or I-Boost-Permutation.

Usage

IBoost(X, Y, data.type, method = "permute", iter.max = 2000,
  v = 0.01, m.stop = 5, alpha.series = c(0.05, seq(0.1, 1, by =
  0.1)), n.fold = 5, permN = 100, seed = 12345678)

Arguments

X	The matrix of all predictors. Each row represents a subject and each column represents a feature (with total of d columns). Each column of X should be standardized.
Y	The survival time, represented by a Surv object.
data.type	The list of indices representing the types of the predictors. Each element of the list is a vector of integers between 1 and d that corresponds to the column numbers of a type of predictors in X. The indices should be non-overlapping.
method	The version of I-Boost to be used; set method=“permute” for I-Boost-Permutation and method=“CV” for I-Boost-CV.
iter.max	The maximum number of iterations.
v	The penalty factor for the current estimate at each iteration.
m.stop	The stopping criterion; if the parameters are not updated consecutively for m.stop iterations, then the algorithm terminates.
alpha.series	The set of values to be considered for the second tuning parameter in elastic net (α) in cross-validation; only applicable when method=”CV”.
n.fold	The number of folds in cross-validation for the selection of tuning parameters; only applicable when method=”CV”.
permN	The number of permutation data sets; only applicable when method="permute".
seed	The initial random seed for partitioning the data set for cross-validation or generating permutation data sets.

Value

A list of two elements, beta and iter.no. beta is a vector of estimated regression parameters. iter.no is the number of iterations used in the estimation.

See Also

glmnet, survival.

alexwky/I-Boost documentation built on May 3, 2019, 5:43 p.m.