Summary of an lbc_net_surv Object

Provides a structured summary of an "lbc_net_surv" object, including training loss, balance assessments, and the estimated survival difference with standard error and confidence interval at a grid of time points.

Usage

# S3 method for class 'lbc_net_surv'
summary(object, ...)

Arguments

object

An object of class "lbc_net_surv", generated by lbc_net_surv().

...

Additional arguments (ignored).

Value

A list containing:

sample_info

Sample sizes and covariate counts.

loss

Training loss.

local_balance

Local standardized differences (LSD) from training.

balance_table

Pre- and post-weighting global standardized differences (GSD).

survival

Estimated survival curves \(S_1(t)\), \(S_0(t)\), survival difference, standard errors, and confidence intervals on the evaluation grid.

gsd

GSD after weighting.

Details

The function extracts key model components using getLBC from the fitted object. It reports:

• Sample size and covariate dimensions;

• Training loss and local balance diagnostics (LSD);

• Global standardized differences (GSD) before and after weighting;

• The estimated survival difference \(\Delta(t) = S_1(t) - S_0(t)\) over the evaluation time grid, including standard errors and Wald 95% confidence intervals.

This method is specifically designed for survival outcomes fitted via lbc_net_surv. For non-survival outcomes, use summary.lbc_net instead.

See also

lbc_net_surv, gsd, getLBC

Examples

if (FALSE) { # \dontrun{
  fit_surv <- lbc_net_surv(
    data    = dat,
    formula = Tr ~ X1 + X2,
    time    = "time",
    delta   = "delta"
  )
  out <- summary(fit_surv)
  names(out)
  head(out$survival$survival_df)
} # }