Calculate Linear Variable Importance of HTEs

Source: R/variable_importance.R

calculate_linear_vimp estimates the linear hypothesis test of removing a particular moderator from a linear model containing all moderators. Unlike calculate_vimp, this will only be unbiased and have correct asymptotic coverage rates if the true model is linear. This linear approach is also substantially faster, so may be useful when prototyping an analysis.

calculate_linear_vimp(
  full_data,
  weight_col,
  pseudo_outcome,
  ...,
  .VIMP_cfg,
  .Model_cfg
)

Arguments

full_data

dataframe

weight_col

Unquoted name of the weight column.

pseudo_outcome

Unquoted name of the pseudo-outcome.

...

Unquoted names of covariates to include in the joint effect model. The variable importance will be calculated for each of these covariates.

.VIMP_cfg

A VIMP_cfg object defining how VIMP should be estimated.

.Model_cfg

A Model_cfg object defining how the joint effect model should be estimated.

References

  • Williamson, B. D., Gilbert, P. B., Carone, M., & Simon, N. (2021). Nonparametric variable importance assessment using machine learning techniques. Biometrics, 77(1), 9-22.

  • Williamson, B. D., Gilbert, P. B., Simon, N. R., & Carone, M. (2021). A general framework for inference on algorithm-agnostic variable importance. Journal of the American Statistical Association, 1-14.

See also

calculate_vimp()