Check to see if weights are optimized correctly

This function checks to see if the optimization is done properly by checking the covariate averages before and after adjustment.

Usage

check_weights(weighted_data, processed_agd)

# S3 method for class 'maicplus_check_weights'
print(
  x,
  mean_digits = 2,
  prop_digits = 2,
  sd_digits = 3,
  digits = getOption("digits"),
  ...
)

Arguments

weighted_data: object returned after calculating weights using estimate_weights
processed_agd: a data frame, object returned after using process_agd or aggregated data following the same naming convention
x: object from check_weights
mean_digits: number of digits for rounding mean columns in the output
prop_digits: number of digits for rounding proportion columns in the output
sd_digits: number of digits for rounding mean columns in the output
digits: minimal number of significant digits, see print.default.
...: further arguments to print.data.frame

Value

data.frame of weighted and unweighted covariate averages of the IPD, average of aggregate data, and sum of inner products of covariate \(x_i\) and the weights (\(exp(x_i\beta)\))

Methods (by generic)

print(maicplus_check_weights): Print method for check_weights objects

Examples

data(weighted_sat)
data(agd)
check_weights(weighted_sat, process_agd(agd))
#>   covariate match_stat internal_trial internal_trial_after_weighted
#> 1       AGE       Mean         59.850                         51.00
#> 2       AGE     Median         59.000                         49.00
#> 3       AGE         SD          9.011                          3.25
#> 4  SEX_MALE       Prop          0.380                          0.49
#> 5     ECOG0       Prop          0.410                          0.35
#> 6     SMOKE       Prop          0.320                          0.19
#>   external_trial sum_centered_IPD_with_weights
#> 1          51.00                        0.0000
#> 2          49.00                        0.0000
#> 3           3.25                       -0.0045
#> 4           0.49                        0.0000
#> 5           0.35                        0.0000
#> 6           0.19                        0.0000