Add an example showing how to update a stratified design at the time analysis (#608)

* Add an example showing how to update a stratified design at the time of analysis

* Fix CI/CD checking errors
    - fix the gt testing
    - remove `pull()` from the example
    - update the Rd file after removing `pull()`
    - add namespace to `group_by` and `summarize`

Author: LittleBeannie

R/gs_update_ahr.R

@@ -118,6 +118,78 @@
 #'
 #' # Alpha is updated to 0.05
 #' gs_update_ahr(x = x, alpha = 0.05)
+#'
+#' # ------------------------------------------------- #
+#' # Two-sided asymmetric stratified design,
+#' # beta-spending with non-binding lower bound
+#' # ------------------------------------------------- #
+#' enroll_rate <- define_enroll_rate(stratum = c("A", "B"), duration = c(12, 12), rate = c(1, 1))
+#'
+#' # We assumme there are 2 strata, "A" and "B".
+#' # For each stratum, there are delayed effect for the first 3 months.
+#' # After the delayed effect, the HR is 0.8 for stratum A and 0.5 for stratum B.
+#' fail_rate <- define_fail_rate(stratum = c("A", "A", "B", "B"),
+#'                               duration = c(3, Inf, 3, Inf),
+#'                               fail_rate = log(2) / c(9, 9, 9, 15),
+#'                               hr = c(1, 0.8, 1, 0.5),
+#'                               dropout_rate = rep(0.001, 4))
+#'
+#' # The original design assumes there are 2 IAs and 1 FA cutting by calendar time.
+#' # The efficacy testing is conducted at IA2 and FA.
+#' # The futility testing is conducted at IA1.
+#' x <- gs_design_ahr(enroll_rate = enroll_rate,
+#'                   fail_rate = fail_rate,
+#'                   alpha = 0.0125,
+#'                   beta = 0.1,
+#'                   analysis = c(20, 28, 36),
+#'                   upper = "gs_spending_bound",
+#'                   upar = list(sf = "sfLDOF", total_spend = 0.0125),
+#'                   lower = "gs_spending_bound",
+#'                   lpar = list(sf = "sfHSD", total_spend = 0.1, param = -8),
+#'                   test_upper = c(FALSE, TRUE, TRUE),
+#'                   test_lower = c(TRUE, FALSE, FALSE)) |> to_integer()
+#'
+#' # At time of analysis
+#' # For IA1,
+#' # - There are 70 events observed during the delayed effect period for stratum A.
+#' # - There are 150 events observed after the delayed effect period for stratum A.
+#' # - There are 75 events observed during the delayed effect period for stratum B.
+#' # - There are 90 events observed after the delayed effect period for stratum B.
+#' # For IA2,
+#' # - There are 75 events observed during the delayed effect period for stratum A.
+#' # - There are 210 events observed after the delayed effect period for stratum A.
+#' # - There are 76 events observed during the delayed effect period for stratum B.
+#' # - There are 136 events observed after the delayed effect period for stratum B.
+#' # For FA,
+#' # - There are 77 events observed during the delayed effect period for stratum A.
+#' # - There are 245 events observed after the delayed effect period for stratum A.
+#' # - There are 77 events observed during the delayed effect period for stratum B.
+#' # - There are 170 events observed after the delayed effect period for stratum B.
+#' event_tbl <- data.frame(analysis = c(1, 1, 1, 1,
+#'                                     2, 2, 2, 2,
+#'                                     3, 3, 3, 3),
+#'                        stratum = c("A", "A", "B", "B", # IA1
+#'                                    "A", "A", "B", "B", # IA2
+#'                                    "A", "A", "B", "B"),# FA
+#'                                  # event per interval per stratum at IA1
+#'                        event = c(70, 150, 75, 90,
+#'                                  # event per interval per stratum at IA2
+#'                                  75, 210, 76, 136,
+#'                                  # event per interval per stratum at FA
+#'                                  77, 245, 77, 170))
+#' observed_event <- (event_tbl |> dplyr::group_by(analysis) |> dplyr::summarize(x = sum(event)))$x
+#'
+#' ustime <- pmin(x$analysis$event,
+#'                observed_event) / x$analysis$event[3]
+#' ustime[3] <- 1
+#' lstime <- ustime
+#'
+#' xu <- gs_update_ahr(x = x,
+#'                     alpha = 0.015,
+#'                     ustime = ustime,
+#'                     lstime = lstime,
+#'                     event_tbl = event_tbl
+#'                     )
 gs_update_ahr <- function(
     x = NULL,
     alpha = NULL,