tests/testthat/_snaps/weights.md

In propensity: A Toolkit for Calculating and Working with Propensity Scores

ATE works for binary cases

Code
  expr
Condition <propensity_range_error>
  Error in `wt_ate()`:
  ! The propensity score must be between 0 and 1.
  i The range of `ps` is -0.1 and 3.3

ATE errors appropriately for categorical with vector propensity scores

Code
  expr
Condition <propensity_matrix_type_error>
  Error:
  ! For categorical exposures, `.propensity` must be a matrix or data frame.

wt_ate() with ps_trim issues refit warning if not refit, no warning if refit

Code
  expr
Condition <propensity_no_refit_warning>
  Warning in `wt_ate()`:
  It appears you trimmed your propensity score but did not refit the model.
  i Use `ps_refit()` for more accurate re-estimation.

Other estimands (att, atu, etc.) with ps_trim or ps_trunc

Code
  expr
Condition <propensity_no_refit_warning>
  Warning in `wt_att()`:
  It appears you trimmed your propensity score but did not refit the model.
  i Use `ps_refit()` for more accurate re-estimation.

wt_atu.ps_trim triggers refit check, sets 'atu; trimmed'

Code
  expr
Condition <propensity_no_refit_warning>
  Warning in `wt_atu()`:
  It appears you trimmed your propensity score but did not refit the model.
  i Use `ps_refit()` for more accurate re-estimation.

wt_atm.ps_trim triggers refit check, sets 'atm; trimmed'

Code
  expr
Condition <propensity_no_refit_warning>
  Warning in `wt_atm()`:
  It appears you trimmed your propensity score but did not refit the model.
  i Use `ps_refit()` for more accurate re-estimation.

wt_ato.ps_trim triggers refit check, sets 'ato; trimmed'

Code
  expr
Condition <propensity_no_refit_warning>
  Warning in `wt_ato()`:
  It appears you trimmed your propensity score but did not refit the model.
  i Use `ps_refit()` for more accurate re-estimation.

wt_entropy works for binary cases

Code
  expr
Condition <propensity_range_error>
  Error in `wt_entropy()`:
  ! The propensity score must be between 0 and 1.
  i The range of `ps` is -0.1 and 3.3

wt_entropy works with ps_trim objects

Code
  expr
Condition <propensity_no_refit_warning>
  Warning in `wt_entropy()`:
  It appears you trimmed your propensity score but did not refit the model.
  i Use `ps_refit()` for more accurate re-estimation.

entropy weights error on unsupported exposure types

Code
  expr
Condition <propensity_matrix_type_error>
  Error:
  ! For categorical exposures, `.propensity` must be a matrix or data frame.

Code
  expr
Condition <propensity_wt_not_supported_error>
  Error in `wt_entropy()`:
  ! Exposure type "continuous" not currently supported for entropy

wt_ate works with data frames

Code
  expr
Condition <propensity_df_ncol_error>
  Error:
  ! `.propensity` data frame must have at least one column.

Code
  expr
Condition <propensity_df_column_error>
  Error:
  ! Column selection failed:

GLM methods error on non-GLM objects

Code
  expr
Condition <propensity_method_error>
  Error in `wt_ate()`:
  ! No method for objects of class character

Code
  expr
Condition <propensity_method_error>
  Error in `wt_att()`:
  ! No method for objects of class list

wt_* functions error appropriately on invalid inputs

Code
  expr
Condition <propensity_range_error>
  Error in `wt_ate()`:
  ! The propensity score must be between 0 and 1.
  i The range of `ps` is -0.1 and 1.1

Code
  expr
Condition <propensity_range_error>
  Error in `wt_att()`:
  ! The propensity score must be between 0 and 1.
  i The range of `ps` is 0.0 and 1.0

Code
  expr
Condition <propensity_length_error>
  Error in `wt_ate()`:
  ! `.propensity` and `.exposure` must have the same length.
  i `.propensity` has length 2
  i `.exposure` has length 3

Code
  expr
Condition <rlang_error>
  Error in `match_exposure_type()`:
  ! `exposure_type` must be one of "auto", "binary", "categorical", or "continuous", not "invalid".

Code
  expr
Condition <propensity_method_error>
  Error in `wt_ate()`:
  ! No method for objects of class character

Code
  expr
Condition <propensity_matrix_type_error>
  Error:
  ! For categorical exposures, `.propensity` must be a matrix or data frame.

data frame methods error appropriately

Code
  expr
Condition <propensity_df_ncol_error>
  Error:
  ! `.propensity` data frame must have at least one column.

Code
  expr
Condition <propensity_df_column_error>
  Error:
  ! Column selection failed:

Code
  expr
Condition <propensity_df_column_error>
  Error:
  ! Column selection failed:

Code
  expr
Condition <simpleWarning>
  Warning in `check_ps_range()`:
  NAs introduced by coercion
Condition <simpleError>
  Error in `.exposure / .propensity`:
  ! non-numeric argument to binary operator

Code
  expr
Condition <propensity_range_error>
  Error in `weight_fn_numeric()`:
  ! The propensity score must be between 0 and 1.
  i The range of `ps` is 0.5 and 1.5

GLM methods error appropriately

Code
  expr
Condition <propensity_method_error>
  Error in `wt_ate()`:
  ! No method for objects of class lm

Code
  expr
Condition <propensity_length_error>
  Error in `wt_ate.numeric()`:
  ! `.propensity` and `.exposure` must have the same length.
  i `.propensity` has length 2
  i `.exposure` has length 4

default methods provide informative errors

Code
  expr
Condition <propensity_method_error>
  Error in `wt_ate()`:
  ! No method for objects of class my_custom_class

Code
  expr
Condition <propensity_method_error>
  Error in `wt_att()`:
  ! No method for objects of class my_custom_class

Code
  expr
Condition <propensity_method_error>
  Error in `wt_atu()`:
  ! No method for objects of class my_custom_class

Code
  expr
Condition <propensity_method_error>
  Error in `wt_atm()`:
  ! No method for objects of class my_custom_class

Code
  expr
Condition <propensity_method_error>
  Error in `wt_ato()`:
  ! No method for objects of class my_custom_class

Code
  expr
Condition <propensity_method_error>
  Error in `wt_entropy()`:
  ! No method for objects of class my_custom_class

GLM methods handle non-binomial families appropriately

Code
  expr
Condition <rlang_error>
  Error in `match_exposure_type()`:
  ! `exposure_type` must be one of "auto", "binary", or "categorical", not "continuous".

all methods handle NAs appropriately

Code
  expr
Condition <propensity_length_error>
  Error in `wt_ate.numeric()`:
  ! `.propensity` and `.exposure` must have the same length.
  i `.propensity` has length 18
  i `.exposure` has length 20

propensity documentation built on March 3, 2026, 1:06 a.m.