GenericML (version 0.2.2)

GenericML_combine: Combine several GenericML objects

Description

This function combines multiple "GenericML" objects into one "GenericML" object. Combining several "GenericML" objects can be useful when you cannot run GenericML() for sufficiently many splits due to memory constraints. In this case, you may run GenericML() multiple times with only a small number of sample splits each and combine the returned "GenericML" objects into one GenericML object with this function.

Usage

GenericML_combine(x)

Arguments

x

A list of "GenericML" objects, as returned by the function GenericML().

Value

A"GenericML" object as returned by GenericML(). In the arguments component of this object, the objects parallel, num_cores, seed, and store_learners are set to NULL as these might differ between the individual GenericML objects in x. Moreover, the propensity_scores component of the returned object is taken from the first "GenericML" object in x.

Details

To ensure consistency of the estimates, all "GenericML" objects in the list x must have the exact same parameter specifications in their original call to GenericML(), except for the parameters num_splits, parallel, num_cores, seed, and store_learners (i.e. these arguments may vary between the "GenericML" objects in the list x). An error will be thrown if this is not satisfied.

See Also

GenericML()

Examples

# NOT RUN {
if (require("glmnet") && require("ranger")) {

## generate data
set.seed(1)
n  <- 150                                  # number of observations
p  <- 5                                    # number of covariates
D  <- rbinom(n, 1, 0.5)                    # random treatment assignment
Z  <- matrix(runif(n*p), n, p)             # design matrix
Y0 <- as.numeric(Z %*% rexp(p) + rnorm(n)) # potential outcome without treatment
Y1 <- 2 + Y0                               # potential outcome under treatment
Y  <- ifelse(D == 1, Y1, Y0)               # observed outcome

## column names of Z
colnames(Z) <- paste0("V", 1:p)

## specify learners
learners <- c("lasso", "mlr3::lrn('ranger', num.trees = 10)")

## glmnet v4.1.3 isn't supported on Solaris, so skip Lasso in this case
if(Sys.info()["sysname"] == "SunOS") learners <- learners[-1]

## call GenericML three times and store the returned objects in a list x
x <- lapply(1:3, function(...) GenericML(Z, D, Y,
                               learners, num_splits = 2,
                               parallel = FALSE))

## combine the objects in x into one GenericML object
genML <- GenericML_combine(x)

## you can use all methods of GenericML objects on the combined object, for instance accessors:
get_BLP(genML, plot = TRUE)
}

# }