Last updated on 2020-05-25 08:50:51 CEST.
| Flavor | Version | Tinstall | Tcheck | Ttotal | Status | Flags |
|---|---|---|---|---|---|---|
| r-devel-linux-x86_64-debian-clang | 2.0.3 | 12.27 | 36.61 | 48.88 | OK | |
| r-devel-linux-x86_64-debian-gcc | 2.0.3 | 11.01 | 28.38 | 39.39 | OK | |
| r-devel-linux-x86_64-fedora-clang | 2.0.3 | 61.58 | NOTE | |||
| r-devel-linux-x86_64-fedora-gcc | 2.0.3 | 60.77 | NOTE | |||
| r-devel-windows-ix86+x86_64 | 2.0.3 | 42.00 | 67.00 | 109.00 | OK | |
| r-patched-linux-x86_64 | 2.0.3 | 10.62 | 36.41 | 47.03 | OK | |
| r-patched-solaris-x86 | 2.0.3 | 91.40 | ERROR | |||
| r-release-linux-x86_64 | 2.0.3 | 11.91 | 36.43 | 48.34 | OK | |
| r-release-osx-x86_64 | 2.0.3 | OK | ||||
| r-release-windows-ix86+x86_64 | 2.0.3 | 44.00 | 85.00 | 129.00 | OK | |
| r-oldrel-osx-x86_64 | 2.0.3 | OK | ||||
| r-oldrel-windows-ix86+x86_64 | 2.0.3 | 41.00 | 81.00 | 122.00 | OK |
Version: 2.0.3
Check: compiled code
Result: NOTE
File ‘norm2/libs/norm2.so’:
Found no calls to: ‘R_registerRoutines’, ‘R_useDynamicSymbols’
It is good practice to register native routines and to disable symbol
search.
See ‘Writing portable packages’ in the ‘Writing R Extensions’ manual.
Flavors: r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc
Version: 2.0.3
Check: examples
Result: ERROR
Running examples in ‘norm2-Ex.R’ failed
The error most likely occurred in:
> ### Name: emNorm
> ### Title: EM algorithm for incomplete multivariate normal data
> ### Aliases: emNorm emNorm.default emNorm.formula emNorm.norm
> ### Keywords: multivariate NA
>
> ### ** Examples
>
> ## run EM for marijuana data with strict convergence criterion
> data(marijuana)
> result <- emNorm(marijuana, criterion=1e-06)
*** caught segfault ***
address 0, cause 'memory not mapped'
Traceback:
1: emNorm.default(marijuana, criterion = 1e-06)
2: emNorm(marijuana, criterion = 1e-06)
An irrecoverable exception occurred. R is aborting now ...
Flavor: r-patched-solaris-x86
Version: 2.0.3
Check: tests
Result: ERROR
Running ‘emTests.R’ [2s/13s]
Running ‘impTests.R’
Comparing ‘impTests.Rout’ to ‘impTests.Rout.save’ ... OK
Running ‘logpostTests.R’
Comparing ‘logpostTests.Rout’ to ‘logpostTests.Rout.save’ ... OK
Running ‘mcmcTests.R’
Comparing ‘mcmcTests.Rout’ to ‘mcmcTests.Rout.save’ ... OK
Running ‘miInferenceTests.R’
Comparing ‘miInferenceTests.Rout’ to ‘miInferenceTests.Rout.save’ ... OK
Running the tests in ‘tests/emTests.R’ failed.
Complete output:
> library(norm2)
>
> ## run EM on fake data with no missing values
> set.seed(1234)
> simdata <- data.frame(
+ Y1=rnorm(6), Y2=rnorm(6), Y3=rnorm(6), X1=rnorm(6) )
> emResult <- emNorm( cbind(Y1,Y2,Y3) ~ X1, data=simdata )
> print( summary( emResult ) )
Predictor (X) variables:
Mean SD Observed Missing Pct.Missing
(Intercept) 1.0000000 0.000000 6 0 0
X1 0.2068512 1.178512 6 0 0
Response (Y) variables:
Mean SD Observed Missing Pct.Missing
Y1 -0.2092854 1.2953550 6 0 0
Y2 -0.6752401 0.2138685 6 0 0
Y3 -0.2141319 0.6864124 6 0 0
Missingness patterns for response (Y) variables
(. denotes observed value, m denotes missing value)
(variable names are displayed vertically)
(rightmost column is the frequency):
YYY
123
... 6
Method: EM
Prior: "uniform"
Convergence criterion: 1e-05
Iterations: 2
Converged: TRUE
Max. rel. difference: 0
-2 Loglikelihood: -8.3665
-2 Log-posterior density: -8.3665
Worst fraction missing information: 0
Estimated coefficients (beta):
Y1 Y2 Y3
(Intercept) -0.3069981 -0.6785479 -0.2457194
X1 0.4723816 0.0159912 0.1527063
Estimated covariance matrix (sigma):
Y1 Y2 Y3
Y1 1.14001811 0.06789367 0.13397509
Y2 0.06789367 0.03782049 0.03942553
Y3 0.13397509 0.03942553 0.36564511
>
> ## impose missing values and run again
> simdata$Y1[3] <- simdata$Y2[4] <- simdata$Y3[4] <- NA
> emResult <- emNorm( cbind(Y1,Y2,Y3) ~ X1, data=simdata )
> print( summary( emResult ) )
Predictor (X) variables:
Mean SD Observed Missing Pct.Missing
(Intercept) 1.0000000 0.000000 6 0 0
X1 0.2068512 1.178512 6 0 0
Response (Y) variables:
Mean SD Observed Missing Pct.Missing
Y1 -0.4680307 1.2630567 5 1 16.66667
Y2 -0.6322806 0.2081665 5 1 16.66667
Y3 -0.2349012 0.7653216 5 1 16.66667
Missingness patterns for response (Y) variables
(. denotes observed value, m denotes missing value)
(variable names are displayed vertically)
(rightmost column is the frequency):
YYY
123
... 4
.mm 1
m.. 1
Method: EM
Prior: "uniform"
Convergence criterion: 1e-05
Iterations: 114
Converged: TRUE
Max. rel. difference: 9.8836e-06
-2 Loglikelihood: -10.70465
-2 Log-posterior density: -10.70465
Worst fraction missing information: 0.9313
Estimated coefficients (beta):
Y1 Y2 Y3
(Intercept) -1.0195738 -0.61863204 -0.1874030
X1 0.5166042 -0.01611584 0.1214564
Estimated covariance matrix (sigma):
Y1 Y2 Y3
Y1 1.80844106 -0.06800034 -0.75502100
Y2 -0.06800034 0.03443290 0.05759078
Y3 -0.75502100 0.05759078 0.41668488
>
> ## redundant Y-variable
> simdata$Y3 <- simdata$Y1 + simdata$Y2
> emResult <- emNorm( cbind(Y1,Y2,Y3) ~ X1, data=simdata )
*** caught segfault ***
address 0, cause 'memory not mapped'
Traceback:
1: emNorm.default(y, x = x, intercept = FALSE, iter.max = iter.max, criterion = criterion, estimate.worst = estimate.worst, starting.values = starting.values, prior = prior, prior.df = prior.df, prior.sscp = prior.sscp, ...)
2: emNorm.formula(cbind(Y1, Y2, Y3) ~ X1, data = simdata)
3: emNorm(cbind(Y1, Y2, Y3) ~ X1, data = simdata)
An irrecoverable exception occurred. R is aborting now ...
Flavor: r-patched-solaris-x86