[Rsiena-commits] r38 - in pkg/RSiena: . man tests

noreply at r-forge.r-project.org noreply at r-forge.r-project.org
Tue Jan 12 19:28:21 CET 2010


Author: ripleyrm
Date: 2010-01-12 19:28:21 +0100 (Tue, 12 Jan 2010)
New Revision: 38

Modified:
   pkg/RSiena/changeLog
   pkg/RSiena/man/RSiena-package.Rd
   pkg/RSiena/man/siena07.Rd
   pkg/RSiena/man/sienaFit.Rd
   pkg/RSiena/man/simstats0c.Rd
   pkg/RSiena/tests/parallel.R
   pkg/RSiena/tests/parallel.Rout.save
Log:
Reduce time taken to run tests and examples.

Modified: pkg/RSiena/changeLog
===================================================================
--- pkg/RSiena/changeLog	2010-01-12 16:45:29 UTC (rev 37)
+++ pkg/RSiena/changeLog	2010-01-12 18:28:21 UTC (rev 38)
@@ -1,3 +1,9 @@
+2010-01-12 R-forge revision 35
+
+	* man/RSiena-package.Rd, man/siena07.Rd, man/sienaFit.Rd,
+	man/simstats0c.Rd: reduce time for examples
+	* tests/parallel.R, tests/patallel.Rout.save: reduce tests
+
 2010-01-02 R-forge revision 34
 
 	* R/sienaprint.r: corrected layout of sienaFitThetaTable for

Modified: pkg/RSiena/man/RSiena-package.Rd
===================================================================
--- pkg/RSiena/man/RSiena-package.Rd	2010-01-12 16:45:29 UTC (rev 37)
+++ pkg/RSiena/man/RSiena-package.Rd	2010-01-12 18:28:21 UTC (rev 38)
@@ -45,9 +45,9 @@
 \references{See \url{http://www.stats.ox.ac.uk/~snijders/siena/}}
 \keyword{ package }
 \examples{
-mynet1 <- sienaNet(array(c(s501, s502, s503), dim=c(50, 50, 3)))
+mynet1 <- sienaNet(array(c(tmp3, tmp4), dim=c(32, 32, 2)))
 mydata <- sienaDataCreate(mynet1)
 myeff <- getEffects(mydata)
-mymodel <- sienaModelCreate(findiff=FALSE, fn=simstats0c)
+mymodel <- sienaModelCreate(findiff=FALSE, fn=simstats0c, nsub=2, n3=100)
 ans <- siena07(mymodel, data=mydata, effects=myeff, batch=TRUE)
 }

Modified: pkg/RSiena/man/siena07.Rd
===================================================================
--- pkg/RSiena/man/siena07.Rd	2010-01-12 16:45:29 UTC (rev 37)
+++ pkg/RSiena/man/siena07.Rd	2010-01-12 18:28:21 UTC (rev 38)
@@ -62,15 +62,16 @@
   objects
   \code{\link{sienaModelCreate}}, \code{\link{print.sienaFit}}}
 \examples{
-mymodel <- sienaModelCreate(fn=simstats0c)
-mynet1 <- sienaNet(array(c(s501, s502, s503), dim=c(50, 50, 3)))
+mymodel <- sienaModelCreate(fn=simstats0c, nsub=2, n3=100)
+mynet1 <- sienaNet(array(c(tmp3, tmp4), dim=c(32, 32, 2)))
 mydata <- sienaDataCreate(mynet1)
 myeff <- getEffects(mydata)
 ans <- siena07(mymodel, data=mydata, effects=myeff, batch=TRUE)
 
 #or for conditional estimation
+\dontrun{
 mymodel$condname <- 'mynet1'
 mymodel$cconditional <- TRUE
-ans <- siena07(mymodel, data=mydata, effects=myeff, batch=TRUE)
+ans <- siena07(mymodel, data=mydata, effects=myeff, batch=TRUE)}
 }
 \keyword{models}

Modified: pkg/RSiena/man/sienaFit.Rd
===================================================================
--- pkg/RSiena/man/sienaFit.Rd	2010-01-12 16:45:29 UTC (rev 37)
+++ pkg/RSiena/man/sienaFit.Rd	2010-01-12 18:28:21 UTC (rev 38)
@@ -51,7 +51,7 @@
  derivative matrix of expected statistics \code{X} by parameters, and the
  covariance matrix of the expected statistics \code{X}.
 
- The function \code{xtable.sientFit} creates an object of class
+ The function \code{xtable.sienaFit} creates an object of class
  \code{xtable.sienaFit} which inherits from class \code{xtable} and
  passes an extra arguments to the \code{print.xtable}.
 }
@@ -62,8 +62,8 @@
 \seealso{\code{\link{xtable}}, \code{\link{print.xtable}},
   \code{\link{siena07}}}
 \examples{
-mymodel <- sienaModelCreate(fn=simstats0c)
-mynet1 <- sienaNet(array(c(s501, s502, s503), dim=c(50, 50, 3)))
+mymodel <- sienaModelCreate(fn=simstats0c, nsub=2, n3=100)
+mynet1 <- sienaNet(array(c(tmp3, tmp4), dim=c(32, 32, 2)))
 mydata <- sienaDataCreate(mynet1)
 myeff <- getEffects(mydata)
 ans <- siena07(mymodel, data=mydata, effects=myeff, batch=TRUE)

Modified: pkg/RSiena/man/simstats0c.Rd
===================================================================
--- pkg/RSiena/man/simstats0c.Rd	2010-01-12 16:45:29 UTC (rev 37)
+++ pkg/RSiena/man/simstats0c.Rd	2010-01-12 18:28:21 UTC (rev 38)
@@ -58,12 +58,12 @@
 \seealso{\code{\link{siena07}} }
 \examples{
 
-mynet1 <- sienaNet(array(c(tmp3,tmp4),dim=c(32,32,2)))
+mynet1 <- sienaNet(array(c(tmp3, tmp4), dim=c(32, 32, 2)))
 mydata <- sienaDataCreate(mynet1)
 myeff<- getEffects(mydata)
 myeff[myeff$effectName=='transitive triplets'&
             myeff$type=='eval','include']<- TRUE
-mymodel<- sienaModelCreate(findiff=TRUE, fn = simstats0c)
+mymodel<- sienaModelCreate(fn = simstats0c, nsub=2, n3=100)
 ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE)
 }
 \keyword{models}

Modified: pkg/RSiena/tests/parallel.R
===================================================================
--- pkg/RSiena/tests/parallel.R	2010-01-12 16:45:29 UTC (rev 37)
+++ pkg/RSiena/tests/parallel.R	2010-01-12 18:28:21 UTC (rev 38)
@@ -1,95 +1,53 @@
-library(RSiena)
-print(packageDescription("RSiena",fields="Repository/R-Forge/Revision"))
+library(RSienaTest)
+print(packageDescription("RSienaTest",fields="Repository/R-Forge/Revision"))
 
-##test1
-print('test1')
-mynet1 <- sienaNet(array(c(s501, s502, s503), dim=c(50, 50, 3)))
-mydata <- sienaDataCreate(mynet1)
-myeff <- getEffects(mydata)
-mymodel<- model.create(findiff=TRUE, fn=simstats0c, projname='test1',
-                       cond=FALSE)
-ans <- siena07(mymodel, data=mydata, effects=myeff, batch=TRUE,parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
-##test2
-print('test2')
-mymodel2 <- mymodel
-mymodel2$cconditional <- TRUE
-mymodel2$condvarno <- 1
-mymodel2$projname <- 'test2'
-ans <- siena07(mymodel2, data=mydata, effects=myeff, batch=TRUE,parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
 ##test3
-mynet1 <- sienaNet(array(c(tmp3,tmp4),dim=c(32,32,2)))
+mynet1 <- sienaNet(array(c(tmp3, tmp4),dim=c(32, 32, 2)))
 mydata <- sienaDataCreate(mynet1)
 myeff<- getEffects(mydata)
 mymodel<- model.create(findiff=TRUE, fn = simstats0c, projname='test3',
-                       cond=FALSE)
+                       cond=FALSE, nsub=2, n3=100)
 print('test3')
-ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
+system.time(ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE))#,dll='../siena/src/RSiena.dll')
 ##test4
 mymodel$projname <- 'test4'
 mymodel$cconditional <- TRUE
 mymodel$condvarno<- 1
 print('test4')
-ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
-##test5
-mynet1 <- sienaNet(array(c(s501, s502, s503), dim=c(50, 50, 3)))
-mydata <- sienaDataCreate(mynet1)
-myeff <- getEffects(mydata)
-mymodel<- model.create(fn=simstats0c, projname='test5',
-                       cond=FALSE)
-print('test5')
-ans <- siena07(mymodel, data=mydata, effects=myeff, batch=TRUE,parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
-##test6
-mymodel2 <- mymodel
-mymodel2$cconditional <- TRUE
-mymodel2$condvarno <- 1
-mymodel2$projname <- 'test6'
-print('test6')
-ans <- siena07(mymodel2, data=mydata, effects=myeff, batch=TRUE,parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
+system.time(ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE))#,dll='../siena/src/RSiena.dll')
 ##test7
 mynet1 <- sienaNet(array(c(tmp3,tmp4),dim=c(32,32,2)))
 mydata <- sienaDataCreate(mynet1)
 myeff<- getEffects(mydata)
-mymodel<- model.create(fn = simstats0c, projname='test7',
+mymodel<- model.create(fn = simstats0c, projname='test7', nsub=2, n3=100,
                        cond=FALSE)
 print('test7')
-ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
+system.time(ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE))#,dll='../siena/src/RSiena.dll')
 ##test8
 mymodel$projname <- 'test8'
 mymodel$cconditional <- TRUE
 mymodel$condvarno<- 1
 print('test8')
-ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
-##test9
+system.time(ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE))#,dll='../siena/src/RSiena.dll')
 
-print('test9')
 mynet1 <- sienaNet(array(c(s501, s502, s503), dim=c(50, 50, 3)))
 mynet2 <- sienaNet(s50a,type='behavior')
 mydata <- sienaDataCreate(mynet1, mynet2)
 myeff <- getEffects(mydata)
-myeff$initialValue[96] <- 0.34699930338 ## siena3 starting values differ
-mymodel<- model.create(findiff=FALSE, fn=simstats0c, projname='test9',
-                       cond=FALSE)
-ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE)
+myeff$initialValue[94] <- 0.34699930338 ## siena3 starting values differ
 ##test10
 print('test10')
 mymodel$projname <- 'test10'
 mymodel$cconditional <- TRUE
 mymodel$condvarno<- 1
-ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE)
+system.time(ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE))
 ##test11
 print('test11')
-data501 <- sienaDataCreateFromSession("s50.csv", modelName="s50")
-data501e <- sienaDataCreateFromSession("s50e.csv", modelName="s50e")
-data501paj <- sienaDataCreateFromSession("s50paj.csv", modelName="s50paj")
+system.time(data501 <- sienaDataCreateFromSession("s50.csv", modelName="s50"))
+system.time(data501e <- sienaDataCreateFromSession("s50e.csv", modelName="s50e"))
+system.time(data501paj <- sienaDataCreateFromSession("s50paj.csv", modelName="s50paj"))
 
-model501 <- model.create( projname="s50",  cond=FALSE)
-model501e <- model.create( projname="s50e", cond=FALSE )
-model501paj <- model.create(projname="s50paj", cond=FALSE )
-ans501 <- siena07(model501, data=data501$mydata, effects=data501$myeff,
-                  parallelTesting=TRUE, batch=TRUE, verbose=TRUE)
-ans501e <- siena07(model501e, data=data501e$mydata, effects=data501e$myeff,
-                   parallelTesting=TRUE, batch=TRUE, verbose=TRUE)
-ans501paj <- siena07(model501paj, data=data501paj$mydata,
-                     effects=data501paj$myeff,
-                  parallelTesting=TRUE, batch=TRUE, verbose=TRUE)
+model501e <- model.create( projname="s50e", cond=FALSE, nsub=2, n3=100 )
+system.time(ans501e <- siena07(model501e, data=data501e$mydata, effects=data501e$myeff,
+                   parallelTesting=TRUE, batch=TRUE, verbose=TRUE))
 ## compare with outputs in parallelchecked/

Modified: pkg/RSiena/tests/parallel.Rout.save
===================================================================
--- pkg/RSiena/tests/parallel.Rout.save	2010-01-12 16:45:29 UTC (rev 37)
+++ pkg/RSiena/tests/parallel.Rout.save	2010-01-12 18:28:21 UTC (rev 38)
@@ -1,6 +1,6 @@
 
-R version 2.9.2 Patched (2009-09-16 r49745)
-Copyright (C) 2009 The R Foundation for Statistical Computing
+R version 2.10.1 Patched (2010-01-11 r50955)
+Copyright (C) 2010 The R Foundation for Statistical Computing
 ISBN 3-900051-07-0
 
 R is free software and comes with ABSOLUTELY NO WARRANTY.
@@ -15,545 +15,59 @@
 'help.start()' for an HTML browser interface to help.
 Type 'q()' to quit R.
 
-> library(RSiena)
-> print(packageDescription("RSiena",fields="Repository/R-Forge/Revision"))
+> library(RSienaTest)
+Loading required package: xtable
+> print(packageDescription("RSienaTest",fields="Repository/R-Forge/Revision"))
 [1] NA
 > 
-> ##test1
-> print('test1')
-[1] "test1"
-> mynet1 <- sienaNet(array(c(s501, s502, s503), dim=c(50, 50, 3)))
+> ##test3
+> mynet1 <- sienaNet(array(c(tmp3, tmp4),dim=c(32, 32, 2)))
 > mydata <- sienaDataCreate(mynet1)
-> myeff <- getEffects(mydata)
-> mymodel<- model.create(findiff=TRUE, fn=simstats0c, projname='test1')
-> ans <- siena07(mymodel, data=mydata, effects=myeff, batch=TRUE,parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
+> myeff<- getEffects(mydata)
+> mymodel<- model.create(findiff=TRUE, fn = simstats0c, projname='test3',
++                        cond=FALSE, nsub=2, n3=100)
+> print('test3')
+[1] "test3"
+> system.time(ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE))#,dll='../siena/src/RSiena.dll')
 
 Stochastic approximation algorithm.
 Initial value for gain parameter = 0.2.
 Start of the algorithm.
 Target function values are 
-  1.   115.0000   2.   106.0000   3.   238.0000   4.   160.0000
+  1.    51.0000   2.   129.0000   3.    58.0000
 
 Start phase 0 
-theta:  4.70  4.33 -1.47  0.00 
+theta:  4.81 -0.56  0.00 
 Current parameter values:
- 4.696042  4.328845 -1.467705  0.000000
+ 4.8094118 -0.5603907  0.0000000
 
 Start phase 1 
 Phase 1 Iteration 1 Progress: 0%
-Phase 1 Iteration 2 Progress: 0%
-Phase 1 Iteration 3 Progress: 0%
-Phase 1 Iteration 4 Progress: 0%
-Phase 1 Iteration 5 Progress: 0%
-Phase 1 Iteration 6 Progress: 0%
-Phase 1 Iteration 7 Progress: 1%
-Phase 1 Iteration 8 Progress: 1%
-Phase 1 Iteration 9 Progress: 1%
-Phase 1 Iteration 10 Progress: 1%
-Phase 1 Iteration 11 Progress: 1%
-Phase 1 Iteration 12 Progress: 1%
-Phase 1 Iteration 13 Progress: 1%
-Phase 1 Iteration 14 Progress: 1%
-Phase 1 Iteration 15 Progress: 1%
-Phase 1 Iteration 16 Progress: 1%
-Phase 1 Iteration 17 Progress: 1%
-Phase 1 Iteration 18 Progress: 1%
-Phase 1 Iteration 19 Progress: 1%
-Time per iteration in phase 1  = 0.07556 
+Phase 1 Iteration 2 Progress: 1%
+Phase 1 Iteration 3 Progress: 1%
+Phase 1 Iteration 4 Progress: 2%
+Phase 1 Iteration 5 Progress: 2%
+Phase 1 Iteration 6 Progress: 3%
+Phase 1 Iteration 7 Progress: 3%
+Phase 1 Iteration 8 Progress: 3%
+Phase 1 Iteration 9 Progress: 4%
+Phase 1 Iteration 10 Progress: 4%
+Phase 1 Iteration 11 Progress: 5%
+Phase 1 Iteration 12 Progress: 5%
+Phase 1 Iteration 13 Progress: 5%
+Phase 1 Iteration 14 Progress: 6%
+Phase 1 Iteration 15 Progress: 6%
+Phase 1 Iteration 16 Progress: 7%
+Time per iteration in phase 1  = 0.0200 
 Average deviations NR generated statistics and targets
 after phase 1:
-      28.105263
-      30.421053
-      23.368421
-    -123.684211
+      32.437500
+       8.687500
+     -25.875000
 
 Diagonal values of derivative matrix :
- 17.1477  16.5353 224.2105  60.0000
+ 18.1935  90.6250  31.2500
 dfra :
- 17.147698   0.000000  29.473684  -9.473684
-  0.000000  16.535344  31.578947  -9.473684
- -3.026064  -2.431668 224.210526  30.526316
- -5.155517  -5.836004  66.315789  60.000000
-
-inverse of dfra :
- 0.0610880666  0.0040772223 -0.0137112006  0.0172651301
- 0.0028583933  0.0647268885 -0.0148890663  0.0182464993
- 0.0001212103 -0.0001740369  0.0052631476 -0.0026860828
- 0.0053930667  0.0068384658 -0.0084435127  0.0228937835
-
-Full Quasi-Newton-Raphson step after phase 1:
-1.     0.614904
-2.     0.555342
-3.    -0.453330
-4.     2.669304 
-This step is multiplied by the factor  0.10000.
-Phase 1 achieved after  19  iterations.
-theta:  4.758  4.384 -1.513  0.267 
-Current parameter values:
- 4.7575328  4.3843796 -1.5130376  0.2669304
-
-Phase 2 has 4 subphases.
-Each subphase can be repeated up to 4 times
-
-Start phase 2.1
-Phase 2 Subphase 1 Iteration 1 Progress: 1%
-Phase 2 Subphase 1 Iteration 2 Progress: 1%
-theta  4.412  3.979 -1.528  0.599 
-ac 3.17 3.07 0.75 1.52 
-Phase 2 Subphase 1 Iteration 3 Progress: 1%
-Phase 2 Subphase 1 Iteration 4 Progress: 2%
-theta  4.17  3.69 -1.59  1.12 
-ac 2.625 3.005 0.854 1.502 
-Phase 2 Subphase 1 Iteration 5 Progress: 2%
-Phase 2 Subphase 1 Iteration 6 Progress: 2%
-theta  4.09  3.60 -1.67  1.46 
-ac 2.568 2.634 0.908 1.495 
-Phase 2 Subphase 1 Iteration 7 Progress: 2%
-Phase 2 Subphase 1 Iteration 8 Progress: 2%
-theta  4.14  3.46 -1.76  1.67 
-ac 1.81 2.40 0.86 1.48 
-Phase 2 Subphase 1 Iteration 9 Progress: 2%
-Phase 2 Subphase 1 Iteration 10 Progress: 2%
-theta  4.58  3.86 -1.83  1.81 
-ac 1.809 2.397 0.848 1.458 
-Phase 2 Subphase 1 Iteration 200 Progress: 5%
-theta  5.74  4.62 -2.38  2.81 
-ac -0.024 -0.116  0.201  0.300 
-Intervention 2.1.1: changes truncated, iterations: 1
-Warning: an autocorrelation is positive at the end of this subphase.
-Autocorrelations:
--0.08107265
--0.06697238
- 0.19914922
- 0.29846032
-
-Time per iteration in phase 2.1 = 0.01555
-theta  5.63  4.41 -2.29  2.67 
-ac -0.081 -0.067  0.199  0.298 
-Phase 2.1 ended after 227 iterations.
-Warning. Autocorrelation criterion not satisfied.
-theta:  5.63  4.41 -2.29  2.67 
-Current parameter values:
- 5.634409  4.408576 -2.291103  2.672469
-
-Start phase 2.2
-Phase 2 Subphase 2 Iteration 1 Progress: 5%
-Phase 2 Subphase 2 Iteration 2 Progress: 5%
-Phase 2 Subphase 2 Iteration 3 Progress: 5%
-Phase 2 Subphase 2 Iteration 4 Progress: 5%
-Phase 2 Subphase 2 Iteration 5 Progress: 5%
-Phase 2 Subphase 2 Iteration 6 Progress: 5%
-Phase 2 Subphase 2 Iteration 7 Progress: 5%
-Phase 2 Subphase 2 Iteration 8 Progress: 5%
-Phase 2 Subphase 2 Iteration 9 Progress: 5%
-Phase 2 Subphase 2 Iteration 10 Progress: 5%
-Time per iteration in phase 2.2 = 0.01629
-theta  5.88  4.52 -2.34  2.80 
-ac -0.17139 -0.04297 -0.00407 -0.12345 
-Phase 2.2 ended after 140 iterations.
-theta:  5.88  4.52 -2.34  2.80 
-Current parameter values:
- 5.884014  4.516232 -2.337510  2.797008
-
-Start phase 2.3
-Phase 2 Subphase 3 Iteration 1 Progress: 9%
-Phase 2 Subphase 3 Iteration 2 Progress: 9%
-Phase 2 Subphase 3 Iteration 3 Progress: 9%
-Phase 2 Subphase 3 Iteration 4 Progress: 9%
-Phase 2 Subphase 3 Iteration 5 Progress: 9%
-Phase 2 Subphase 3 Iteration 6 Progress: 9%
-Phase 2 Subphase 3 Iteration 7 Progress: 9%
-Phase 2 Subphase 3 Iteration 8 Progress: 9%
-Phase 2 Subphase 3 Iteration 9 Progress: 9%
-Phase 2 Subphase 3 Iteration 10 Progress: 9%
-Phase 2 Subphase 3 Iteration 200 Progress: 12%
-theta  5.88  4.50 -2.36  2.87 
-ac -0.0113  0.0289 -0.0913 -0.2268 
-Time per iteration in phase 2.3 = 0.0162
-theta  5.88  4.56 -2.36  2.82 
-ac  0.0267  0.0707 -0.0909 -0.1898 
-Phase 2.3 ended after 371 iterations.
-theta:  5.88  4.56 -2.36  2.82 
-Current parameter values:
- 5.876732  4.557273 -2.359675  2.820789
-
-Start phase 2.4
-Phase 2 Subphase 4 Iteration 1 Progress: 15%
-Phase 2 Subphase 4 Iteration 2 Progress: 15%
-Phase 2 Subphase 4 Iteration 3 Progress: 15%
-Phase 2 Subphase 4 Iteration 4 Progress: 15%
-Phase 2 Subphase 4 Iteration 5 Progress: 15%
-Phase 2 Subphase 4 Iteration 6 Progress: 15%
-Phase 2 Subphase 4 Iteration 7 Progress: 15%
-Phase 2 Subphase 4 Iteration 8 Progress: 15%
-Phase 2 Subphase 4 Iteration 9 Progress: 15%
-Phase 2 Subphase 4 Iteration 10 Progress: 15%
-Phase 2 Subphase 4 Iteration 200 Progress: 18%
-theta  5.91  4.56 -2.36  2.81 
-ac  0.0513  0.1045  0.0121 -0.0514 
-Phase 2 Subphase 4 Iteration 400 Progress: 21%
-theta  5.99  4.58 -2.35  2.82 
-ac  0.01805  0.05127  0.04037 -0.00067 
-Phase 2 Subphase 4 Iteration 600 Progress: 24%
-theta  5.88  4.56 -2.35  2.83 
-ac 0.0408 0.0264 0.0433 0.0345 
-Warning: an autocorrelation is positive at the end of this subphase.
-Autocorrelations:
-0.03663243
-0.02358606
-0.03208522
-0.05905858
-
-Time per iteration in phase 2.4 = 0.01498
-theta  5.89  4.51 -2.36  2.82 
-ac 0.0366 0.0236 0.0321 0.0591 
-Phase 2.4 ended after 630 iterations.
-Warning. Autocorrelation criterion not satisfied.
-theta:  5.89  4.51 -2.36  2.82 
-Current parameter values:
- 5.891445  4.509223 -2.356034  2.819908
-
-Start phase 3 
-Simulated values, phase 3.
-Phase 3 Iteration 50 Progress 28%
-Phase 3 Iteration 100 Progress 32%
-Phase 3 Iteration 150 Progress 36%
-Phase 3 Iteration 200 Progress 39%
-Phase 3 Iteration 250 Progress 43%
-Phase 3 Iteration 300 Progress 47%
-Phase 3 Iteration 350 Progress 51%
-Phase 3 Iteration 400 Progress 54%
-Phase 3 Iteration 450 Progress 58%
-Phase 3 Iteration 500 Progress 62%
-Phase 3 Iteration 550 Progress 66%
-Phase 3 Iteration 600 Progress 70%
-Phase 3 Iteration 650 Progress 73%
-Phase 3 Iteration 700 Progress 77%
-Phase 3 Iteration 750 Progress 81%
-Phase 3 Iteration 800 Progress 85%
-Phase 3 Iteration 850 Progress 89%
-Phase 3 Iteration 900 Progress 92%
-Phase 3 Iteration 950 Progress 96%
-Phase 3 Iteration 1000 Progress 100%
-Time per iteration in phase 3   =  0.08661 
-dfrac :
-  9.7571520   0.0000000  25.1200000  -1.6600000
-  0.0000000  12.3589538  24.3000000   4.5500000
-  0.4003371   1.0164373 221.7000000  93.3500000
- -0.0809192   1.4969350 169.8200000 114.4000000
-
-inverse of dfra :
- 0.1041458815 -0.0007574245 -0.0343988346  0.0296106613
- 0.0008034366  0.0810137195 -0.0173662475  0.0109603189
--0.0005823038  0.0002040916  0.0121798651 -0.0099552941
- 0.0009275484 -0.0013635699 -0.0178773783  0.0233968290
-
-A full Quasi-Newton-Raphson step after phase 3
-would add the following numbers to the parameters, yielding the following results:
-         change     new value 
-   1.    -0.041395    5.850050
-   2.     0.029326    4.538549
-   3.    -0.002098   -2.358132
-   4.     0.003855    2.823763 
-
-unconditional moment estimation.
-Information for convergence diagnosis.
-Averages, standard deviations, and t-ratios for deviations from targets:
-  1.   0.4630   9.4253   0.0491
-  2.  -0.3290   8.6415  -0.0381
-  3.   0.0920  15.3955   0.0060
-  4.  -0.1320  14.2377  -0.0093
-
-Total of 2387 iteration steps.
-
- at 3
-Estimates and standard errors
-                             
- 1. rate:  constant mynet1 rate (period 1)               5.8914  (   0.9350)
- 2. rate:  constant mynet1 rate (period 2)               4.5092  (   0.6868)
- 3. eval:  outdegree (density)                          -2.3560  (   0.0985)
- 4. eval:  reciprocity                                   2.8199  (   0.1744)
-
-Derivative matrix of expected statistics X by parameters:
-
-   9.7571520   0.0000000  25.1200000  -1.6600000
-  0.0000000  12.3589538  24.3000000   4.5500000
-  0.4003371   1.0164373 221.7000000  93.3500000
- -0.0809192   1.4969350 169.8200000 114.4000000
-
-Covariance matrix of X (correlations below the diagonal):
-    88.835     20.485     29.754      7.931
-     0.252     74.675     21.098      8.005
-     0.205      0.159    237.023    186.897
-     0.059      0.065      0.853    202.713
-
-
-> ##test2
-> print('test2')
-[1] "test2"
-> mymodel2 <- mymodel
-> mymodel2$cconditional <- TRUE
-> mymodel2$condvarno <- 1
-> mymodel2$projname <- 'test2'
-> ans <- siena07(mymodel2, data=mydata, effects=myeff, batch=TRUE,parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
-
-Stochastic approximation algorithm.
-Initial value for gain parameter = 0.2.
-Start of the algorithm.
-Target function values are 
-  1.   238.0000   2.   160.0000
-
-Start phase 0 
-theta: -1.47  0.00 
-Current parameter values:
--1.467705  0.000000
-
-Start phase 1 
-Phase 1 Iteration 1 Progress: 0%
-Phase 1 Iteration 2 Progress: 0%
-Phase 1 Iteration 3 Progress: 0%
-Phase 1 Iteration 4 Progress: 0%
-Phase 1 Iteration 5 Progress: 0%
-Phase 1 Iteration 6 Progress: 0%
-Phase 1 Iteration 7 Progress: 0%
-Phase 1 Iteration 8 Progress: 1%
-Phase 1 Iteration 9 Progress: 1%
-Phase 1 Iteration 10 Progress: 1%
-Phase 1 Iteration 11 Progress: 1%
-Phase 1 Iteration 12 Progress: 1%
-Phase 1 Iteration 13 Progress: 1%
-Time per iteration in phase 1  = 0.0325 
-Average deviations NR generated statistics and targets
-after phase 1:
-      15.692308
-    -107.692308
-
-Diagonal values of derivative matrix :
-210.7692  55.3846
-dfra :
-210.76923  60.00000
- 84.61538  55.38462
-
-inverse of dfra :
- 0.008396125 -0.009095802
--0.012827413  0.031951920
-
-Full Quasi-Newton-Raphson step after phase 1:
-1.    -1.111302
-2.     3.642268 
-This step is multiplied by the factor  0.10000.
-Phase 1 achieved after  13  iterations.
-theta: -1.579  0.364 
-Current parameter values:
--1.5788348  0.3642268
-
-Phase 2 has 4 subphases.
-Each subphase can be repeated up to 4 times
-
-Start phase 2.1
-Phase 2 Subphase 1 Iteration 1 Progress: 1%
-Phase 2 Subphase 1 Iteration 2 Progress: 1%
-theta -1.579  0.711 
-ac 0.00 1.45 
-Phase 2 Subphase 1 Iteration 3 Progress: 1%
-Phase 2 Subphase 1 Iteration 4 Progress: 1%
-theta -1.64  1.18 
-ac 0.00 1.30 
-Phase 2 Subphase 1 Iteration 5 Progress: 1%
-Phase 2 Subphase 1 Iteration 6 Progress: 1%
-theta -1.68  1.52 
-ac 0.378 1.111 
-Phase 2 Subphase 1 Iteration 7 Progress: 1%
-Phase 2 Subphase 1 Iteration 8 Progress: 1%
-theta -1.75  1.77 
-ac 0.625 1.070 
-Phase 2 Subphase 1 Iteration 9 Progress: 1%
-Phase 2 Subphase 1 Iteration 10 Progress: 1%
-theta -1.83  1.89 
-ac 0.736 1.041 
-Phase 2 Subphase 1 Iteration 200 Progress: 6%
-theta -2.39  2.87 
-ac 0.00234 0.09035 
-Time per iteration in phase 2.1 = 0.01473
-theta -2.31  2.72 
-ac -0.0383  0.0489 
-Phase 2.1 ended after 222 iterations.
-theta: -2.31  2.72 
-Current parameter values:
--2.314309  2.723011
-
-Start phase 2.2
-Phase 2 Subphase 2 Iteration 1 Progress: 6%
-Phase 2 Subphase 2 Iteration 2 Progress: 6%
-Phase 2 Subphase 2 Iteration 3 Progress: 6%
-Phase 2 Subphase 2 Iteration 4 Progress: 6%
-Phase 2 Subphase 2 Iteration 5 Progress: 6%
-Phase 2 Subphase 2 Iteration 6 Progress: 6%
-Phase 2 Subphase 2 Iteration 7 Progress: 6%
-Phase 2 Subphase 2 Iteration 8 Progress: 6%
-Phase 2 Subphase 2 Iteration 9 Progress: 6%
-Phase 2 Subphase 2 Iteration 10 Progress: 6%
-Phase 2 Subphase 2 Iteration 200 Progress: 11%
-theta -2.39  2.88 
-ac 0.1031 0.0382 
-Warning: an autocorrelation is positive at the end of this subphase.
-Autocorrelations:
-0.106880509
-0.001288962
-
-Time per iteration in phase 2.2 = 0.01447
-theta -2.38  2.85 
-ac 0.10688 0.00129 
-Phase 2.2 ended after 255 iterations.
-Warning. Autocorrelation criterion not satisfied.
-theta: -2.38  2.85 
-Current parameter values:
--2.375661  2.845126
-
-Start phase 2.3
-Phase 2 Subphase 3 Iteration 1 Progress: 12%
-Phase 2 Subphase 3 Iteration 2 Progress: 12%
-Phase 2 Subphase 3 Iteration 3 Progress: 12%
-Phase 2 Subphase 3 Iteration 4 Progress: 12%
-Phase 2 Subphase 3 Iteration 5 Progress: 12%
-Phase 2 Subphase 3 Iteration 6 Progress: 12%
-Phase 2 Subphase 3 Iteration 7 Progress: 12%
-Phase 2 Subphase 3 Iteration 8 Progress: 12%
-Phase 2 Subphase 3 Iteration 9 Progress: 12%
-Phase 2 Subphase 3 Iteration 10 Progress: 12%
-Time per iteration in phase 2.3 = 0.01471
-theta -2.38  2.89 
-ac -0.256 -0.227 
-Phase 2.3 ended after 138 iterations.
-theta: -2.38  2.89 
-Current parameter values:
--2.381897  2.886809
-
-Start phase 2.4
-Phase 2 Subphase 4 Iteration 1 Progress: 19%
-Phase 2 Subphase 4 Iteration 2 Progress: 19%
-Phase 2 Subphase 4 Iteration 3 Progress: 19%
-Phase 2 Subphase 4 Iteration 4 Progress: 19%
-Phase 2 Subphase 4 Iteration 5 Progress: 20%
-Phase 2 Subphase 4 Iteration 6 Progress: 20%
-Phase 2 Subphase 4 Iteration 7 Progress: 20%
-Phase 2 Subphase 4 Iteration 8 Progress: 20%
-Phase 2 Subphase 4 Iteration 9 Progress: 20%
-Phase 2 Subphase 4 Iteration 10 Progress: 20%
-Phase 2 Subphase 4 Iteration 200 Progress: 24%
-theta -2.39  2.85 
-ac -0.1250 -0.0958 
-Time per iteration in phase 2.4 = 0.01441
-theta -2.38  2.87 
-ac -0.1492 -0.0243 
-Phase 2.4 ended after 347 iterations.
-theta: -2.38  2.87 
-Current parameter values:
--2.384249  2.868362
-
-Start phase 3 
-Simulated values, phase 3.
-Phase 3 Iteration 100 Progress 39%
-Phase 3 Iteration 200 Progress 45%
-Phase 3 Iteration 300 Progress 52%
-Phase 3 Iteration 400 Progress 59%
-Phase 3 Iteration 500 Progress 66%
-Phase 3 Iteration 600 Progress 73%
-Phase 3 Iteration 700 Progress 80%
-Phase 3 Iteration 800 Progress 86%
-Phase 3 Iteration 900 Progress 93%
-Phase 3 Iteration 1000 Progress 100%
-Time per iteration in phase 3   =  0.04675 
-dfrac :
-206.42  98.96
-157.82 116.70
-
-inverse of dfra :
- 0.01377585 -0.01168173
--0.01862986  0.02436685
-
-A full Quasi-Newton-Raphson step after phase 3
-would add the following numbers to the parameters, yielding the following results:
-         change     new value 
-   1.     0.002063   -2.382186
-   2.    -0.000887    2.867475 
-
-conditional moment estimation.
-Information for convergence diagnosis.
-Averages, standard deviations, and t-ratios for deviations from targets:
-  1.  -0.3380  15.3236  -0.0221
-  2.  -0.2220  14.3792  -0.0154
-
-Total of 1975 iteration steps.
-
- at 3
-Estimates and standard errors
-                             
-Rate parameters:
- 0.1  Rate parameter period 1                 5.7636  (   0.9461)
- 0.2  Rate parameter period 2                 4.4706  (   0.6605)
-
-Other parameters:
- 1. eval:  outdegree (density)                          -2.3842  (   0.1123)
- 2. eval:  reciprocity                                   2.8684  (   0.1859)
-
-Derivative matrix of expected statistics X by parameters:
-
- 206.42  98.96
-157.82 116.70
-
-Covariance matrix of X (correlations below the diagonal):
-   234.813    186.908
-     0.848    206.761
-
-
-> ##test3
-> mynet1 <- sienaNet(array(c(tmp3,tmp4),dim=c(32,32,2)))
-> mydata <- sienaDataCreate(mynet1)
-> myeff<- getEffects(mydata)
-> mymodel<- model.create(findiff=TRUE, fn = simstats0c, projname='test3')
-> print('test3')
-[1] "test3"
-> ans<- siena07(mymodel, data=mydata, effects=myeff,  batch=TRUE, parallelTesting=TRUE, verbose=TRUE)#,dll='../siena/src/RSiena.dll')
-
-Stochastic approximation algorithm.
-Initial value for gain parameter = 0.2.
-Start of the algorithm.
-Target function values are 
-  1.    51.0000   2.   129.0000   3.    58.0000
-
-Start phase 0 
-theta:  4.81 -0.56  0.00 
-Current parameter values:
- 4.8094118 -0.5603907  0.0000000
-
-Start phase 1 
-Phase 1 Iteration 1 Progress: 0%
-Phase 1 Iteration 2 Progress: 0%
-Phase 1 Iteration 3 Progress: 0%
-Phase 1 Iteration 4 Progress: 0%
-Phase 1 Iteration 5 Progress: 0%
-Phase 1 Iteration 6 Progress: 0%
-Phase 1 Iteration 7 Progress: 1%
-Phase 1 Iteration 8 Progress: 1%
-Phase 1 Iteration 9 Progress: 1%
-Phase 1 Iteration 10 Progress: 1%
-Phase 1 Iteration 11 Progress: 1%
-Phase 1 Iteration 12 Progress: 1%
-Phase 1 Iteration 13 Progress: 1%
-Phase 1 Iteration 14 Progress: 1%
-Phase 1 Iteration 15 Progress: 1%
-Phase 1 Iteration 16 Progress: 1%
-Time per iteration in phase 1  = 0.02467 
-Average deviations NR generated statistics and targets
-after phase 1:
-      32.437500
-       8.687500
-     -25.875000
-
-Diagonal values of derivative matrix :
- 18.1935  90.6250  31.2500
-dfra :
 18.193493  4.375000  1.875000
  4.938234 90.625000 34.375000
 -3.638699 50.000000 31.250000
@@ -573,28 +87,28 @@
 Current parameter values:
  4.6292124 -0.6194468  0.1563076
 
-Phase 2 has 4 subphases.
+Phase 2 has 2 subphases.
 Each subphase can be repeated up to 4 times
 
 Start phase 2.1
-Phase 2 Subphase 1 Iteration 1 Progress: 1%
-Phase 2 Subphase 1 Iteration 2 Progress: 1%
+Phase 2 Subphase 1 Iteration 1 Progress: 7%
+Phase 2 Subphase 1 Iteration 2 Progress: 7%
 theta  4.244 -0.648  0.310 
 ac 1.46 1.30 1.00 
-Phase 2 Subphase 1 Iteration 3 Progress: 1%
-Phase 2 Subphase 1 Iteration 4 Progress: 1%
+Phase 2 Subphase 1 Iteration 3 Progress: 7%
+Phase 2 Subphase 1 Iteration 4 Progress: 7%
 theta  3.772 -0.694  0.540 
 ac 1.459 1.480 0.836 
-Phase 2 Subphase 1 Iteration 5 Progress: 1%
-Phase 2 Subphase 1 Iteration 6 Progress: 1%
+Phase 2 Subphase 1 Iteration 5 Progress: 7%
+Phase 2 Subphase 1 Iteration 6 Progress: 7%
 theta  3.42 -0.76  0.72 
 ac 1.25 1.94 0.65 
-Phase 2 Subphase 1 Iteration 7 Progress: 1%
-Phase 2 Subphase 1 Iteration 8 Progress: 1%
+Phase 2 Subphase 1 Iteration 7 Progress: 7%
+Phase 2 Subphase 1 Iteration 8 Progress: 8%
 theta  3.244 -0.783  0.912 
 ac 0.799 0.794 0.653 
-Phase 2 Subphase 1 Iteration 9 Progress: 1%
-Phase 2 Subphase 1 Iteration 10 Progress: 1%
+Phase 2 Subphase 1 Iteration 9 Progress: 8%
+Phase 2 Subphase 1 Iteration 10 Progress: 8%
 theta  2.969 -0.838  1.040 
 ac 0.792 0.780 0.653 
 Warning: an autocorrelation is positive at the end of this subphase.
@@ -603,7 +117,7 @@
 0.1117977
 0.1982289
 
-Time per iteration in phase 2.1 = 0.004667
+Time per iteration in phase 2.1 = 0.004089
 theta  3.10 -1.09  1.67 
 ac 0.121 0.112 0.198 
 Phase 2.1 ended after 225 iterations.
@@ -613,17 +127,17 @@
  3.102896 -1.092006  1.668237
 
 Start phase 2.2
-Phase 2 Subphase 2 Iteration 1 Progress: 5%
-Phase 2 Subphase 2 Iteration 2 Progress: 5%
-Phase 2 Subphase 2 Iteration 3 Progress: 5%
-Phase 2 Subphase 2 Iteration 4 Progress: 5%
-Phase 2 Subphase 2 Iteration 5 Progress: 5%
-Phase 2 Subphase 2 Iteration 6 Progress: 5%
-Phase 2 Subphase 2 Iteration 7 Progress: 5%
-Phase 2 Subphase 2 Iteration 8 Progress: 5%
-Phase 2 Subphase 2 Iteration 9 Progress: 5%
-Phase 2 Subphase 2 Iteration 10 Progress: 5%
-Time per iteration in phase 2.2 = 0.004449
+Phase 2 Subphase 2 Iteration 1 Progress: 30%
+Phase 2 Subphase 2 Iteration 2 Progress: 31%
+Phase 2 Subphase 2 Iteration 3 Progress: 31%
+Phase 2 Subphase 2 Iteration 4 Progress: 31%
+Phase 2 Subphase 2 Iteration 5 Progress: 31%
+Phase 2 Subphase 2 Iteration 6 Progress: 31%
+Phase 2 Subphase 2 Iteration 7 Progress: 31%
+Phase 2 Subphase 2 Iteration 8 Progress: 31%
+Phase 2 Subphase 2 Iteration 9 Progress: 31%
+Phase 2 Subphase 2 Iteration 10 Progress: 31%
+Time per iteration in phase 2.2 = 0.004030
 theta  3.03 -1.13  1.79 
 ac  0.0471 -0.0488  0.0117 
 Phase 2.2 ended after 263 iterations.
@@ -631,107 +145,63 @@
 Current parameter values:
  3.026403 -1.134321  1.792116
 
-Start phase 2.3
-Phase 2 Subphase 3 Iteration 1 Progress: 10%
-Phase 2 Subphase 3 Iteration 2 Progress: 10%
-Phase 2 Subphase 3 Iteration 3 Progress: 10%
-Phase 2 Subphase 3 Iteration 4 Progress: 10%
-Phase 2 Subphase 3 Iteration 5 Progress: 10%
-Phase 2 Subphase 3 Iteration 6 Progress: 10%
-Phase 2 Subphase 3 Iteration 7 Progress: 10%
-Phase 2 Subphase 3 Iteration 8 Progress: 10%
-Phase 2 Subphase 3 Iteration 9 Progress: 10%
-Phase 2 Subphase 3 Iteration 10 Progress: 10%
-Time per iteration in phase 2.3 = 0.004881
-theta  3.08 -1.15  1.78 
-ac -0.0061 -0.1404 -0.1409 
-Phase 2.3 ended after 168 iterations.
-theta:  3.08 -1.15  1.78 
-Current parameter values:
- 3.077319 -1.153246  1.782989
-
-Start phase 2.4
-Phase 2 Subphase 4 Iteration 1 Progress: 17%
-Phase 2 Subphase 4 Iteration 2 Progress: 17%
-Phase 2 Subphase 4 Iteration 3 Progress: 17%
[TRUNCATED]

To get the complete diff run:
    svnlook diff /svnroot/rsiena -r 38


More information about the Rsiena-commits mailing list