Robustness Value — robustness_value • senseweight

Returns the estimated robustness value, for a specified proportion change

Usage

robustness_value(estimate, b_star = 0, sigma2, weights)

Arguments

estimate: Weighted estimate
b_star: Threshold that corresponds to a substantively meaningful change to the research conclusion. For example, a value of 0 denotes that the bias from omitting a variable was sufficiently large to change the estimate to zero.
sigma2: Estimated variance of the outcome (i.e., stats::var(Y) for obervational setting; stats::var(tau) for generalization setting)
weights: Vector of estimated weights

Value

Robustness value for a specified proportion change

Examples

data(jtpa_women)
site_name <- "NE"
df_site <- jtpa_women[which(jtpa_women$site == site_name), ]
df_else <- jtpa_women[which(jtpa_women$site != site_name), ]

# Estimate unweighted estimator:
model_dim <- estimatr::lm_robust(Y ~ T, data = df_site)
PATE <- coef(lm(Y ~ T, data = df_else))[2]
DiM <- coef(model_dim)[2]
# Generate weights using observed covariates:
df_all <- jtpa_women
df_all$S <- ifelse(jtpa_women$site == "NE", 1, 0)
model_ps <- WeightIt::weightit(
  (1 - S) ~ . - site - T - Y, 
  data = df_all, method = "ebal", estimand = "ATT"
)
weights <- model_ps$weights[df_all$S == 1]
# Estimate IPW model:
model_ipw <- estimatr::lm_robust(Y ~ T, data = df_site, weights = weights)
ipw <- coef(model_ipw)[2]
# Estimate bound for var(tau):
vartau <- var(df_site$Y[df_site$T == 1]) - var(df_site$Y[df_site$T == 0])
RV <- robustness_value(estimate = ipw, b_star = 0, sigma2 = vartau, weights = weights)

print(RV)
#> [1] 0.4113622