[Vinecopula-commits] r70 - in pkg: R inst
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Fr Nov 21 17:37:24 CET 2014
Author: ulf
Date: 2014-11-21 17:37:23 +0100 (Fri, 21 Nov 2014)
New Revision: 70
Modified:
pkg/R/BiCopSelect.r
pkg/R/gof_PIT.r
pkg/inst/ChangeLog
Log:
Bug fix
Modified: pkg/R/BiCopSelect.r
===================================================================
--- pkg/R/BiCopSelect.r 2014-10-28 09:49:10 UTC (rev 69)
+++ pkg/R/BiCopSelect.r 2014-11-21 16:37:23 UTC (rev 70)
@@ -240,7 +240,7 @@
optiout[[37]] = MLE_intern(cbind(data1,data2),start[[37]],37,weights=weights)
if(optiout[[37]]$par[1] >= -0.1 | optiout[[37]]$par[2] >= -1.1){
if(optiout[[37]]$par[1] >= -0.1){
- todo[todo==37] = 24
+ todo[todo==37] = 34
todo = unique(todo)
}else if(optiout[[37]]$par[2] >= -1.1){
todo[todo==37] = 33
@@ -254,10 +254,10 @@
optiout[[38]] = MLE_intern(cbind(data1,data2),start[[38]],38,weights=weights)
if(optiout[[38]]$par[1] >= -1.1 | optiout[[38]]$par[2] >= -1.1){
if(optiout[[38]]$par[1] >= -1.1){
- todo[todo==38] = 24
+ todo[todo==38] = 34
todo = unique(todo)
}else if(optiout[[38]]$par[2] >= -1.1){
- todo[todo==38] = 26
+ todo[todo==38] = 36
todo = unique(todo)
}
optiout[[38]] = list()
Modified: pkg/R/gof_PIT.r
===================================================================
--- pkg/R/gof_PIT.r 2014-10-28 09:49:10 UTC (rev 69)
+++ pkg/R/gof_PIT.r 2014-11-21 16:37:23 UTC (rev 70)
@@ -105,8 +105,10 @@
{
S[i]=sum(tmp[i,])
}
- if(statisticName==1 && method==1) # Macht nur Sinn bei Breymann
+ if(statisticName==1 && method==1){ # Macht nur Sinn bei Breymann
+ require(ADGofTest)
pvalue=ad.test(S,pchisq,df=d)$p.value
+ }
else if(statisticName==2 && method==1)
pvalue=ks.test(S,"pchisq",df=d)$p.value
else
Modified: pkg/inst/ChangeLog
===================================================================
--- pkg/inst/ChangeLog 2014-10-28 09:49:10 UTC (rev 69)
+++ pkg/inst/ChangeLog 2014-11-21 16:37:23 UTC (rev 70)
@@ -12,6 +12,7 @@
* Bootstrap procedure for the PIT based and the ECP based test were incorrect.
First, C starts to count at 0 not 1. This could result in zero entries in the bootstrapped data matrix.
Second, forget to permute vdirect and vindirect according to the permutation of data.
+ * BiCopSelect: For the rotated BB7 and BB8 (family=37, 38) the limiting cases were incorrect for very small parameters (copy&paste error) (Reported by Radek Solnicky. Thanks!)
Version 1.3 (March 26, 2014)
Mehr Informationen über die Mailingliste Vinecopula-commits