RMST fixed design

[fb-elong]

Please refer example in Roxygen2 to check if it meet the goal.

One issue I had is about the enrollRate and failRate.

Should I update the value in enrollRate and failRate after calculating the sample size? In gsDesign, the valuate is treated as ratio instead of actual enrollment rate, because the sample size is determined by power.

I saw you update Power in your example. It confuse me whether you are calculating sample size or power.

> y <- fixed_design("AHR", 
+                   alpha = .025, Power = .9, 
+                   enrollRates = tibble::tibble(Stratum = "All",  duration = 18, rate = 1),
+                   failRates = tibble::tibble(Stratum = "All", duration = c(4, 100), failRate = log(2) / 12, hr = c(1, .6), dropoutRate = .001),
+                   studyDuration = 36)

$analysis
# A tibble: 1 × 7
  Design     N Events  Time Bound alpha  Power
  <chr>  <dbl>  <dbl> <dbl> <dbl> <dbl>  <dbl>
1 AHR       18   12.6    36  1.96 0.025 0.0928