Processes SHAP values in one-hot (wide) format to create beta coefficient corrections.
This includes:
scaling shap values of continuous variables by the predictor value for that row
migrating shap values to the bias for continuous variables where the predictor value was zero
migrating shap values to the bias for categorical variables where the predictor value was reference level
Usage
beta_corrections_derive(
shap_wide,
wide_input_frame,
iblm_model,
migrate_reference_to_bias = TRUE
)Arguments
- shap_wide
Data frame containing SHAP values from XGBoost that have been converted to wide format by [shap_to_onehot()]
- wide_input_frame
Wide format input data frame (one-hot encoded).
- iblm_model
Object of class 'iblm'
- migrate_reference_to_bias
Logical, migrate the beta corrections to the bias for reference levels? This applied to categorical vars only. It is recommended to leave this setting on TRUE
Examples
df_list <- freMTPLmini |>
dplyr::mutate(LogExposure = log(Exposure), .keep = "unused") |>
split_into_train_validate_test(seed = 9000)
iblm_model <- train_iblm_xgb(
df_list,
response_var = "ClaimNb",
offset_var = "LogExposure",
family = "poisson"
)
shap <- extract_booster_shap(iblm_model$booster_model, df_list$test)
wide_input_frame <- data_to_onehot(df_list$test, iblm_model)
shap_wide <- shap_to_onehot(shap, wide_input_frame, iblm_model)
beta_corrections <- beta_corrections_derive(shap_wide, wide_input_frame, iblm_model)
beta_corrections |> dplyr::glimpse()
#> Rows: 3,764
#> Columns: 17
#> $ bias <dbl> -0.03010190, -0.03010190, -0.03010190, -0.03010190, -0.030…
#> $ BonusMalus <dbl> -1.425127e-04, 9.072397e-04, -8.565751e-04, -1.274934e-03,…
#> $ DrivAge <dbl> -7.341999e-03, 3.301016e-03, 2.897860e-03, 3.001865e-03, 3…
#> $ VehAge <dbl> 0.011503229, -0.270234734, 0.014044377, 0.009927072, 0.006…
#> $ VehPower <dbl> -0.0023957209, -0.0059312492, 0.0037696779, 0.0034830798, …
#> $ AreaA <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
#> $ AreaB <dbl> 0.00000000, 0.00000000, 0.00000000, 0.00000000, 0.00000000…
#> $ AreaC <dbl> 0.000000000, 0.143781751, 0.000000000, 0.000000000, 0.0000…
#> $ AreaD <dbl> -0.0044879219, 0.0000000000, 0.0000000000, -0.0023310741, …
#> $ AreaE <dbl> 0.000000e+00, 0.000000e+00, 1.612203e-02, 0.000000e+00, 1.…
#> $ VehBrandB1 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
#> $ VehBrandB12 <dbl> 0.000000000, -0.153895795, 0.000000000, 0.000000000, -0.02…
#> $ VehBrandB2 <dbl> 0.00000000, 0.00000000, 0.01185244, -0.03065922, 0.0000000…
#> $ VehBrandB3 <dbl> 0.00000000, 0.00000000, 0.00000000, 0.00000000, 0.00000000…
#> $ VehBrandB4 <dbl> -0.07181698, 0.00000000, 0.00000000, 0.00000000, 0.0000000…
#> $ VehBrandB5 <dbl> 0.00000000, 0.00000000, 0.00000000, 0.00000000, 0.00000000…
#> $ VehBrandB6 <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.0…