[Gsdesign-commits] r346 - in pkg/gsDesign: . R

[noreply at r-forge.r-project.org]

Author: keaven
Date: 2013-04-20 14:03:08 +0200 (Sat, 20 Apr 2013)
New Revision: 346

Modified:
   pkg/gsDesign/DESCRIPTION
   pkg/gsDesign/R/gsBinomial.R
Log:
Added capability for nSurv, gsSurv, eEvents to accomodate 0 event rates and censoring rates in same period (enables studies with follow-up time cut off). Fixed to nBinomial for relative risk sample size null hypothesis under superiority.

Modified: pkg/gsDesign/DESCRIPTION
===================================================================
--- pkg/gsDesign/DESCRIPTION	2013-03-14 21:41:24 UTC (rev 345)
+++ pkg/gsDesign/DESCRIPTION	2013-04-20 12:03:08 UTC (rev 346)
@@ -1,5 +1,5 @@
 Package: gsDesign
-Version: 2.7-06
+Version: 2.8-00
 Title: Group Sequential Design
 Author: Keaven Anderson 
 Maintainer: Keaven Anderson <keaven_anderson at merck.com>

Modified: pkg/gsDesign/R/gsBinomial.R
===================================================================
--- pkg/gsDesign/R/gsBinomial.R	2013-03-14 21:41:24 UTC (rev 345)
+++ pkg/gsDesign/R/gsBinomial.R	2013-04-20 12:03:08 UTC (rev 346)
@@ -218,7 +218,7 @@
         p10 <- (-b - sqrt(b ^ 2 - 4 * a * c)) / 2 / a
         p20 <- p10 / RR
         p10[d0] <- (p1[d0] + ratio[d0] * p2[d0]) / (1 + ratio[d0])
-        p20[d0] <- p1[d0]
+        p20[d0] <- p10[d0]
         sigma0 <- sqrt((ratio + 1) * 
                         (p10 * (1 - p10) + RR ^ 2 * p20 * (1 - p20) / ratio))
         sigma1 <- sqrt((ratio + 1) *