ps_pond_one {adapt4pv}R Documentation

weihting on propensity score for one drug exposure

Description

Implement the weighting on propensity score with Matching Weights (MW) or the Inverse Probability of Treatment Weighting (IPTW) for one drug exposure. The binary outcome is regressed on the drug exposure of interest through a classical weighted regression. Internal function, not supposed to be used directly.

Usage

ps_pond_one(
  ps_est,
  y,
  weights_type = c("mw", "iptw"),
  truncation = FALSE,
  q = 0.025
)

Arguments

ps_est

An object of class "ps", "*" where "*" is "bic", "hdps" or "xgb" according on how the score was estimated, respective outputs of internal functions est_ps_bic, est_ps_hdps, est_ps_xgb. It is a list with the following elements : * score_type: character, name of the drug exposure for which the PS was estimated. * indicator_expo: indicator of the drugs exposure for which the PS was estimated. One-column Matrix object. * score_variables: Character vector, names of covariate(s) selected to include in the PS estimation model. Could be empty. *score: One-column Matrix object, the estimated score.

y

Binary response variable, numeric.

weights_type

Character. Indicates which type of weighting is implemented. Could be either "mw" or "iptw".

truncation

Bouleen, should we do weight truncation? Default is FALSE.

q

If truncation is TRUE, quantile value for weight truncation. Ignored if truncation is FALSE. Default is 2.5 \%.

Details

The MW are defined by

mw_i = min(PS_i, 1-PS_i)/[(expo_i) * PS_i + (1-expo_i) * (1-PS_i) ]

and weights from IPTW by

iptw_i = expo_i/PS_i + (1-expo_i)/(1-PS_i)

where expo_i is the drug exposure indicator. The PS could be estimated in different ways: using lasso-bic approach, the hdPS algorithm or gradient tree boosting using functions est_ps_bic, est_ps_hdps and est_ps_xgb respectivelly.

Value

An object with S3 class "ps","*" , where "*" is "mw" or "iptw", same as the input parameter weights_type

expo_name

Character, name of the drug exposure for which the PS was estimated.

estimate

Regression coefficient associated with the drug exposure in adjustment on PS.

pval_1sided

One sided p-value associated with the drug exposure in adjustment on PS.

pval_2sided

Two sided p-value associated with the drug exposure in adjustment on PS.

Could return NA if the adjustment on the PS did not converge.

Author(s)

Emeline Courtois
Maintainer: Emeline Courtois emeline.courtois@inserm.fr

Examples


set.seed(15)
drugs <- matrix(rbinom(100*20, 1, 0.2), nrow = 100, ncol = 20)
colnames(drugs) <- paste0("drugs",1:ncol(drugs))
ae <- rbinom(100, 1, 0.3)
pshdps2 <- est_ps_hdps(idx_expo = 2, x = drugs, y = ae, keep_total = 10)
pondps2 <- ps_pond_one(ps_est = pshdps2, y = ae, weights_type = "iptw")
pondps2$estimate #estimated strength of association between drug_2 and the outcome by PS weighting


[Package adapt4pv version 0.2-1 Index]