Auto-Bin a table of a continuous moderating variable into a discrete moderating variable

Use a regression tree to optimally bin a continuous variable, this function will print out a table with estimates and 95

Usage

table_moderator_c_bin(.model, moderator, .name = "bin")

Arguments

.model

a model produced by `bartCause::bartc()`

moderator

the moderator as a vector

.name

sting representing the name of the moderating variable

Value

a data.frame object

Author

George Perrett

Examples

# \donttest{
data(lalonde)
confounders <- c('age', 'educ', 'black', 'hisp', 'married', 'nodegr')
model_results <- bartCause::bartc(
 response = lalonde[['re78']],
 treatment = lalonde[['treat']],
 confounders = as.matrix(lalonde[, confounders]),
 estimand = 'ate',
 commonSuprule = 'none'
)
#> fitting treatment model via method 'bart'
#> fitting response model via method 'bart'
table_moderator_c_bin(model_results, lalonde$age, .name = 'age')
#> Joining with `by = join_by(splits)`
#> Adding missing grouping variables: `moderator`
#> # A tibble: 4 × 5
#> # Groups:   moderator [4]
#>   moderator   est    sd     lci   uci
#>   <chr>     <dbl> <dbl>   <dbl> <dbl>
#> 1 age:17-19 1005.  809.  -636.  2552.
#> 2 age:20-24 1401.  720.   -62.8 2751.
#> 3 age:25-42 1999.  722.   634.  3458.
#> 4 age:43-55 1318. 1440. -2295.  3794.
# }