pseudo_outcome_generator_gbm

Cross-fitting is applied when generate the target pseudo-outcome. This function uses Gradient Boosting to conduct model training on folds then estimates \(E[Y_{t+1}|H_t, A_t], E[A_t|H_t], E[A_t|S_t]\) for the reserved fold. Then the function returns a dataset with column named "yDR" for DR-WCLS pseudo outcome.

Usage

pseudo_outcome_generator_gbm(
  fold,
  ID,
  data,
  Ht,
  St,
  At,
  prob,
  outcome,
  core_num = NULL
)

Arguments

fold

number of folds to split when do corss-fitting

ID

the name of column where participants' ID are stored

data

dataset name

Ht

a vector that contains column names of control variables

St

a vector that contains column names of moderator variables; St should be a subset of Ht

At

column names of treatment (At)

prob

column names of \(p_t(A_t = 1|H_t)\), the experiment design treatment probability

outcome

column names of outcome variable

core_num

number of cores will be used for calculation

Value

This function returns a dataset with pseudo outcome. It learns appropriate working models with Gradient Boosting and generates pseudo outcome using the DR-WCLS.

Examples

 sim_data = generate_dataset(N = 1000, T = 40, P = 50, sigma_residual = 1.5, sigma_randint = 1.5, main_rand = 3, rho = 0.7,
  beta_logit = c(-1, 1.6 * rep(1/50, 50)), model = ~ state1 + state2 + state3 + state4,
  beta = matrix(c(-1, 1.7, 1.5, -1.3, -1),ncol = 1),
  theta1 = 0.8)

Ht = unlist(lapply(1:50, FUN = function(X) paste0("state",X)))
St = unlist(lapply(1:25, FUN = function(X) paste0("state",X)))

pseudo_outcome_generator_gbm(fold = 5,ID = "id", sim_data, Ht, St, "action", "outcome",core_num = 5)
#> Error in pseudo_outcome_generator_gbm(fold = 5, ID = "id", sim_data, Ht,     St, "action", "outcome", core_num = 5): argument "outcome" is missing, with no default