remove AHR examples and will save them in a vignette

Author: LittleBeannie

R/gs_cp_npe.R

@@ -43,84 +43,12 @@
 #' library(gsDesign2)
 #' library(dplyr)
 #'
-#' # ---------------------------------- #
-#' #             Example 1              #
-#' # CP under arbitrary theta and info  #
-#' # ---------------------------------- #
+#' # Calculate conditional power under arbitrary theta and info
+#' # In practice, the value of theta and info commonly comes from a design.
+#' # More examples are available at the pkgdown vignettes.
 #' gs_cp_npe(theta = c(.1, .2),
 #'           info = c(15, 35),
 #'           a = 1.5, b = 1.96)
-#'
-#' # ---------------------------------- #
-#' #             Example 2              #
-#' # Calculate conditional power and    #
-#' #       error of a design            #
-#' # ---------------------------------- #
-#' x <- gs_design_ahr(
-#'   fail_rate = define_fail_rate(duration = c(4, Inf),
-#'                                fail_rate = log(2) / 10,
-#'                                hr = c(1, 0.7),
-#'                                dropout_rate = 0.0001),
-#'   analysis_time = c(12, 24, 36))
-#'
-#' # Example 2A ----
-#' # Conditional error of FA given IA1 Z-score of 1.75
-#' gs_cp_npe(theta = c(0, 0),
-#'           info = x$analysis$info0[c(1, 3)],
-#'           a = 1.75,
-#'           b = x$bound$z[x$bound$bound == "upper" & x$bound$analysis == 3])
-#'
-#' # Example 2B ----
-#' # Conditional power of FA given
-#' # (1) IA1 Z-score of 1.75;
-#' # (2) H1 assumed treatment effect
-#' gs_cp_npe(theta = x$analysis$theta[c(1, 3)],
-#'           # taking info0 in this example gives minor differences
-#'           info = x$analysis$info[c(1, 3)],
-#'           a = 1.75,
-#'           b = x$bound$z[x$bound$bound == "upper" & x$bound$analysis == 3])
-#'
-#' # Example 2C ----
-#' # Assume at IA1:
-#' # (1) The Z-score is 1.75;
-#' # (2) There are 50 events observed during the first 4 months since study starts
-#' # (3) There are 150 events observed after the 4th month.
-#' # For IA1, we take the blinded estimation of theta and info.
-#' # For FA, we take the planned theta and info.
-#' theta_blinded <- -sum(log(c(1, 0.7)) * c(50, 150)) / 200
-#' info_blinded <- 200 / 4
-#' gs_cp_npe(theta = c(theta_blinded, x$analysis$theta[3]),
-#'           info = c(info_blinded, x$analysis$info0[3]),
-#'           a = 1.75,
-#'           b = x$bound$z[x$bound$bound == "upper" & x$bound$analysis == 3])
-#'
-#' # Example 2D ----
-#' # If the HR is not assumed, say, HR = 1 for the first 4 months and 0.8 afterwards
-#' # At FA, assume there are 70 events during the first 4 months and 700 events afterwards.
-#' # We first calculate the expected events at IA1 under the new HR
-#' e_event_ia1 <- expected_event(enroll_rate = x$enroll_rate,
-#'                               fail_rate = x$fail_rate |> mutate(hr = c(1, 0.8)),
-#'                               total_duration = x$analysis$time[1],
-#'                               simple = FALSE)
-#' # Under the new HR, there are 150 events during the first 4 months and 67 events afterwards.
-#' # If 160 and 67 are the blinded events, we can derive the blinded estimation of treatment effect.
-#' theta_blinded_ia1 <- -sum(log(c(1, 0.8)) * c(150, 67)) / (150 + 67)
-#' info_blinded_ia1 <- (150 + 67) / 4
-#'
-#' # We further calculate the expected events at FA under the new HR
-#' e_event_fa <- expected_event(enroll_rate = x$enroll_rate,
-#'                              fail_rate = x$fail_rate |> mutate(hr = c(1, 0.8)),
-#'                              total_duration = x$analysis$time[3],
-#'                              simple = FALSE)
-#' # Under the new HR, there are 223 events during the first 4 months and 562 events afterwards.
-#' # If 160 and 67 are the blinded events, we can derive the blinded estimation of treatment effect.
-#' theta_blinded_fa <- -sum(log(c(1, 0.8)) * c(223, 562)) / (223 + 562)
-#' info_blinded_fa <- (223 + 562) / 4
-#'
-#' gs_cp_npe(theta = c(theta_blinded_ia1, theta_blinded_fa),
-#'           info = c(info_blinded_ia1, info_blinded_fa),
-#'           a = 1.75,
-#'           b = x$bound$z[x$bound$bound == "upper" & x$bound$analysis == 3])
 gs_cp_npe <- function(theta = NULL,
                       info = NULL,
                       a = NULL, b = NULL
@@ -140,15 +68,8 @@ gs_cp_npe <- function(theta = NULL,
     stop("Please provide info (statistical information given the treatment effect theta) to calculate conditional power.")
   }
 
-  if (max(info) > max_info) {
-    stop("The max(info) should be smaller than max_info.")
-  }
-
   check_info(info)
 
-  # calculate information fraction
-  info_frac <- info / max_info
-
   # ----------------------------------------- #
   #   calculate conditional power under theta #
   # ----------------------------------------- #