Hotfix for bcf.predict without propensity scores

jaredsmurray · web-flow · commit 472147f448dd · 2025-10-23T11:55:22.000-05:00
the new dataset for predictions takes X_combined, which wasn't created when propensity_covariate="none"
diff --git a/R/bcf.R b/R/bcf.R
@@ -2707,6 +2707,7 @@ predict.bcfmodel <- function(
     }
 
     # Add propensities to covariate set if necessary
+    X_combined <- X
     if (object$model_params$propensity_covariate != "none") {
         X_combined <- cbind(X, propensity)
     }

Original file line number	Diff line number	Diff line change
`@@ -2707,6 +2707,7 @@ predict.bcfmodel <- function(`
`2707`	`2707`	`}`
`2708`	`2708`
`2709`	`2709`	`# Add propensities to covariate set if necessary`
	`2710`	`+ X_combined <- X`
`2710`	`2711`	`if (object$model_params$propensity_covariate != "none") {`
`2711`	`2712`	`X_combined <- cbind(X, propensity)`
`2712`	`2713`	`}`