intxsurv: Estimate the CATE model using specified scoring methods for survival outcomes

Description

Coefficients of the CATE estimated with random forest, boosting, naive Poisson, two regression, and contrast regression

Usage

intxsurv(
  y,
  d,
  trt,
  x.cate,
  x.ps,
  x.ipcw,
  yf = NULL,
  tau0,
  surv.min = 0.025,
  score.method = c("randomForest", "boosting", "poisson", "twoReg", "contrastReg"),
  ps.method = "glm",
  minPS = 0.01,
  maxPS = 0.99,
  ipcw.method = "breslow",
  initial.predictor.method = "randomForest",
  tree.depth = 3,
  n.trees.rf = 1000,
  n.trees.boosting = 150,
  B = 3,
  Kfold = 5,
  plot.gbmperf = TRUE,
  error.maxNR = 0.001,
  max.iterNR = 100,
  tune = c(0.5, 2),
  ...
)

Value

Depending on what score.method is, the outputs is a combination of the following: result.randomForest: Results of random forest fit, for trt = 0 and trt = 1 separately result.boosting: Results of boosting fit, for trt = 0 and trt = 1 separately result.poisson: Naive Poisson estimator (beta1 - beta0); vector of length p.cate + 1 result.twoReg: Two regression estimator (beta1 - beta0); vector of length p.cate + 1 result.contrastReg: A list of the contrast regression results with 2 elements: $delta.contrastReg: Contrast regression DR estimator; vector of length p.cate + 1 $converge.contrastReg: Indicator that the Newton Raphson algorithm converged for delta_0; boolean

Arguments

y: Observed survival or censoring time; vector of size n.
d: The event indicator, normally 1 = event, 0 = censored; vector of size n.
trt: Treatment received; vector of size n with treatment coded as 0/1.
x.cate: Matrix of p.cate baseline covariates specified in the outcome model; dimension n by p.cate.
x.ps: Matrix of p.ps baseline covariates specified in the propensity score model; dimension n by p.ps.
x.ipcw: Matrix of p.ipw baseline covariate specified in inverse probability of censoring weighting; dimension n by p.ipw.
yf: Follow-up time, interpreted as the potential censoring time; vector of size n if the potential censoring time is known.
tau0: The truncation time for defining restricted mean time lost.
surv.min: Lower truncation limit for probability of being censored (positive and very close to 0).
score.method: A vector of one or multiple methods to estimate the CATE score. Allowed values are: 'randomForest', 'boosting', 'poisson', 'twoReg', 'contrastReg'. Default specifies all 5 methods.
ps.method: A character vector for the method to estimate the propensity score. Allowed values include one of: 'glm' for logistic regression with main effects only (default), or 'lasso' for a logistic regression with main effects and LASSO penalization on two-way interactions (added to the model if interactions are not specified in ps.model). Relevant only when ps.model has more than one variable.
minPS: A numerical value (in `[0, 1]`) below which estimated propensity scores should be truncated. Default is 0.01.
maxPS: A number above which estimated propensity scores should be trimmed; scalar
ipcw.method: The censoring model. Allowed values are: 'breslow' (Cox regression with Breslow estimator of the baseline survivor function), 'aft (exponential)', 'aft (weibull)', 'aft (lognormal)' or 'aft (loglogistic)'. Default is 'breslow'.
initial.predictor.method: A character vector for the method used to get initial outcome predictions conditional on the covariates in cate.model in score.method = 'twoReg' and 'contrastReg'. Allowed values include one of 'randomForest', 'boosting' and 'logistic' (fastest). Default is 'randomForest'.
tree.depth: A positive integer specifying the depth of individual trees in boosting (usually 2-3). Used only if score.method = 'boosting' or if score.method = 'twoReg' or 'contrastReg' and initial.predictor.method = 'boosting'. Default is 3.
n.trees.rf: A positive integer specifying the maximum number of trees in random forest. Used if score.method = 'ranfomForest' or if initial.predictor.method = 'randomForest' with score.method = 'twoReg' or 'contrastReg'. Default is 1000.
n.trees.boosting: A positive integer specifying the maximum number of trees in boosting (usually 100-1000). Used if score.method = 'boosting' or if initial.predictor.method = 'boosting' with score.method = 'twoReg' or 'contrastReg'. Default is 150.
B: A positive integer specifying the number of time cross-fitting is repeated in score.method = 'twoReg' and 'contrastReg'. Default is 3.
Kfold: A positive integer specifying the number of folds (parts) used in cross-fitting to partition the data in score.method = 'twoReg' and 'contrastReg'. Default is 5.
plot.gbmperf: A logical value indicating whether to plot the performance measures in boosting. Used only if score.method = 'boosting' or if score.method = 'twoReg' or 'contrastReg' and initial.predictor.method = 'boosting'. Default is TRUE.
error.maxNR: A numerical value > 0 indicating the minimum value of the mean absolute error in Newton Raphson algorithm. Used only if score.method = 'contrastReg'. Default is 0.001.
max.iterNR: A positive integer indicating the maximum number of iterations in the Newton Raphson algorithm. Used only if score.method = 'contrastReg'. Default is 100.
tune: A vector of 2 numerical values > 0 specifying tuning parameters for the Newton Raphson algorithm. tune[1] is the step size, tune[2] specifies a quantity to be added to diagonal of the slope matrix to prevent singularity. Used only if score.method = 'contrastReg'. Default is c(0.5, 2).
...: Additional arguments for gbm()