varying trial size and refs

Oncostat · Jul 6, 2017 · 26cd3df · 26cd3df
1 parent ec9ac30
commit 26cd3df
Show file tree

Hide file tree

Showing 6 changed files with 43 additions and 27 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -2,8 +2,8 @@ Package: surrosurv
 Type: Package
 Title: Evaluation of Failure Time Surrogate Endpoints in Individual Patient Data
     Meta-Analyses
-Version: 1.1.20
-Date: 2017-06-21
+Version: 1.1.22
+Date: 2017-06-26
 Authors@R: c(
   person("Federico", "Rotolo", email="[email protected]", 
           role=c("aut", "cre")),
@@ -16,11 +16,12 @@ Author: Federico Rotolo [aut, cre],
 Maintainer: Federico Rotolo <[email protected]>
 Description: Provides functions for the evaluation of
     surrogate endpoints when both the surrogate and the true endpoint are failure
-    time variables. The approaches implemented are: (1) the two-step approach
-    (Burzykowski et al, 2001) <DOI:10.1111/1467-9876.00244> with a copula model (Clayton, Plackett, Hougaard) at
+    time variables. The approaches implemented are:
+    (1) the two-step approach (Burzykowski et al, 2001) <DOI:10.1111/1467-9876.00244> with a copula model (Clayton, Plackett, Hougaard) at
     the first step and either a linear regression of log-hazard ratios at the second
-    step (either adjusted or not for measurement error); (2) mixed proportional
-    hazard models estimated via mixed Poisson GLM.
+    step (either adjusted or not for measurement error);
+    (2) mixed proportional hazard models estimated via mixed Poisson GLM
+    (Rotolo et al, 2017 <DOI:10.1177/0962280217718582>).
 Depends: R (>= 2.10), stats, optimx, grDevices
 Imports:
     copula,

diff --git a/R/loocv.R b/R/loocv.R
@@ -96,17 +96,16 @@ loocv.data.frame <- function(object,
     redobject$trialref <- factor(redobject$trialref)
     predint <- tryCatch(
       expr = {
-        surrofit <-
-          surrosurv(data = redobject, models = models2predict, ...)
-        res <-
-          lapply(attr(predict(surrofit), 'predf'), function(x)
-            x(alpha))
+        surrofit <- surrosurv(data = redobject, models = models2predict, ...)
+        res <- lapply(attr(predict(surrofit), 'predf'), function(x) x(alpha))
         names <- names(res)
         res <- lapply(1:length(res), function(i)
           rbind(res[[i]], t(print(
             surrofit, silent = TRUE
           ))[, i, drop = FALSE]))
         names(res) <- names
+        attr(res, 'convergence') <- convergence(surrofit)
+        attr(res, 'convals') <- convals(surrofit)
         res
       },
       error = function(e) {
@@ -124,18 +123,23 @@ loocv.data.frame <- function(object,
           models2predict <- c(models2predict[0:(poipos - 1)],
                               paste0('Poisson', c('T', 'TI', 'TIa')))
         }
-        return(mapply(function(i)
+        res <- mapply(function(i)
           return(t(t(c(fit = NA,
                        lwr = NA,
                        upr = NA,
                        kTau = NA,
                        R2 = NA)
           ))),
-          models2predict, SIMPLIFY = FALSE))
+          models2predict, SIMPLIFY = FALSE)
+        attr(res, 'convergence') <- attr(res, 'convals') <- 
+          matrix(NA, length(models2predict), 3)
+        return(res)
       }
     )
     RES <- c(list(margPars = c(alpha = alpha, beta = beta)),
              predint)
+    attr(RES, 'convergence') <- attr(predint, 'convergence')
+    attr(RES, 'convals') <- attr(predint, 'convals')
     return(RES)
   }
 
@@ -147,8 +151,7 @@ loocv.data.frame <- function(object,
   }
   clusterEvalQ(cl, library('survival'))
   clusterEvalQ(cl, library('surrosurv'))
-  loocvRES <-
-    clusterApplyLB(cl, levels(object$trialref), loof, ...)
+  loocvRES <- clusterApplyLB(cl, levels(object$trialref), loof, ...)
   stopCluster(cl)
   rm(cl)
 

diff --git a/R/simData.R b/R/simData.R
@@ -12,7 +12,7 @@ kTaus <- Vectorize(function(lnSigma2)
 # plot(exp(logsigma2), kTaus_ln(logsigma2))
 
 find.sigma2 <- function(tau) {
-  loss <- function(x)
+  loss <- function(x) 
     (kTaus(x) - tau) ^ 2
   RES <- optimize(loss, c(-10, 4))
   return(exp(RES$minimum))
@@ -49,9 +49,9 @@ simData <- function(method = c('re', 'cc', 'mx')) {
   ) {
 
     if (length(gammaWei) == 1)
-      gammaWei <- rep(gammaWei, 2) 
+      gammaWei <- rep(gammaWei, 2)
 
-    if (!nifix) ni <- round(runif(N, ni * .75, ni * 1.25))
+    if (!nifix) ni <- round(runif(N, ni * .5, ni * 1.5))
     trialref <- unlist(mapply(rep, 1:N, each = ni))
 
     data <- data.frame(trialref = factor(trialref),

diff --git a/man/simData.Rd b/man/simData.Rd
@@ -125,10 +125,18 @@ simData.mx(R2 = 0.6, N = 30, ni = 200,
 Burzykowski T, Cortinas Abrahantes J (2005). Validation in the case of two failure-time endpoints.
 In \emph{The Evaluation of Surrogate Endpoints} (pp. 163-194). Springer, New York.
 
+Rotolo F, Paoletti X, Burzykowski T, Buyse M, Michiels S.
+  A Poisson approach for the validation of failure time surrogate endpoints
+  in individual patient data meta-analyses.
+  \emph{Statistical Methhods in Medical Research} 2017;
+  \bold{In Press}.
+  \doi{10.1177/0962280217718582}
+  
 Shi Q, Renfro LA, Bot BM, Burzykowski T, Buyse M, Sargent DJ.
 Comparative assessment of trial-level surrogacy measures for candidate
 time-to-event surrogate endpoints in clinical trials. 
-\emph{Computational Statistics & Data Analysis} 2011; 55: 2748--2757.
+\emph{Computational Statistics & Data Analysis} 2011; 
+\bold{55}: 2748--2757.
 }
 
 \author{

diff --git a/man/surrosurv-package.Rd b/man/surrosurv-package.Rd
@@ -20,15 +20,17 @@
 }
 \references{
   Rotolo F, Paoletti X, Burzykowski T, Buyse M, Michiels S.
-  Evaluation of failure time surrogate endpoints in individual patient data meta-analyses
-  of randomized clinical trials. A Poisson approach.
-  \emph{Working paper}.
+  A Poisson approach for the validation of failure time surrogate endpoints
+  in individual patient data meta-analyses.
+  \emph{Statistical Methhods in Medical Research} 2017;
+  \bold{In Press}.
+  \doi{10.1177/0962280217718582}
 
   Burzykowski T, Molenberghs G, Buyse M et al.
   Validation of surrogate end points in multiple randomized clinical trials with failure time end points.
   \emph{Journal of the Royal Statistical Society C} 2001;
   \bold{50}:405--422.
-  \doi{10.1111/1467-9876.00244}.
+  \doi{10.1111/1467-9876.00244}
 
   Gasparrini A, Armstrong B, Kenward MG.
   Multivariate meta-analysis for non-linear and other multi-parameter associations.

diff --git a/man/surrosurv.Rd b/man/surrosurv.Rd
@@ -192,10 +192,12 @@ surrosurv(data,
   \emph{Journal of the America Statistical Association} 1965; \bold{60}:516--522.
   \doi{10.1080/01621459.1965.10480807}
 
-  Rotolo F, Paoletti X, Burzykowski T, Buyse M, Michiels S.
-  Evaluation of failure time surrogate endpoints in individual patient data
-  meta-analyses of randomized clinical trials. A Poisson approach.
-  \emph{Working paper}.
+ Rotolo F, Paoletti X, Burzykowski T, Buyse M, Michiels S.
+  A Poisson approach for the validation of failure time surrogate endpoints
+  in individual patient data meta-analyses.
+  \emph{Statistical Methhods in Medical Research} 2017;
+  \bold{In Press}.
+  \doi{10.1177/0962280217718582}
 
   Shih JH, Louis TA.
   Inferences on the Association Parameter in Copula Models