CRAN Package Check Results for Package emIRT

Last updated on 2019-06-25 19:49:38 CEST.

Flavor Version Tinstall Tcheck Ttotal Status Flags
r-devel-linux-x86_64-debian-clang 0.0.8 814.62 43.93 858.55 OK
r-devel-linux-x86_64-debian-gcc 0.0.8 619.97 34.49 654.46 OK
r-devel-linux-x86_64-fedora-clang 0.0.8 1449.52 NOTE
r-devel-linux-x86_64-fedora-gcc 0.0.8 1251.59 NOTE
r-devel-windows-ix86+x86_64 0.0.8 2062.00 114.00 2176.00 NOTE
r-patched-linux-x86_64 0.0.8 742.92 43.46 786.38 OK
r-patched-solaris-x86 0.0.8 1080.00 ERROR
r-release-linux-x86_64 0.0.8 742.07 42.98 785.05 OK
r-release-windows-ix86+x86_64 0.0.8 1567.00 143.00 1710.00 NOTE
r-release-osx-x86_64 0.0.8 ERROR
r-oldrel-windows-ix86+x86_64 0.0.8 1379.00 104.00 1483.00 NOTE
r-oldrel-osx-x86_64 0.0.8 ERROR

Check Details

Version: 0.0.8
Check: installed package size
Result: NOTE
     installed size is 17.4Mb
     sub-directories of 1Mb or more:
     data 2.0Mb
     libs 15.2Mb
Flavors: r-devel-linux-x86_64-fedora-clang, r-devel-windows-ix86+x86_64, r-release-windows-ix86+x86_64, r-oldrel-windows-ix86+x86_64

Version: 0.0.8
Check: compiled code
Result: NOTE
    File ‘emIRT/libs/emIRT.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: 0.0.8
Check: examples
Result: ERROR
    Running examples in ‘emIRT-Ex.R’ failed
    The error most likely occurred in:
    
    > ### Name: ordIRT
    > ### Title: Two-parameter Ordinal IRT estimation via EM
    > ### Aliases: ordIRT
    > ### Keywords: multivariate
    >
    > ### ** Examples
    >
    >
    > ## Not run:
    > ##D ### Real data example: Asahi-Todai survey (not run)
    > ##D ## Collapses 5-category ordinal survey items into 3 categories for estimation
    > ##D data(AsahiTodai)
    > ##D out.varinf <- ordIRT(.rc = AsahiTodai$dat.all, .starts = AsahiTodai$start.values,
    > ##D .priors = AsahiTodai$priors, .D = 1,
    > ##D .control = {list(verbose = TRUE,
    > ##D thresh = 1e-6, maxit = 500)})
    > ##D
    > ##D ## Compare against MCMC estimates using 3 and 5 categories
    > ##D cor(ideal3, out.varinf$means$x)
    > ##D cor(ideal5, out.varinf$means$x)
    > ## End(Not run)
    >
    >
    > ### Monte Carlo simulation of ordIRT() model vs. known parameters
    > ## Set number of legislators and items
    > set.seed(2)
    > NN <- 500
    > JJ <- 100
    >
    > ## Simulate true parameters from original model
    > x.true <- runif(NN, -2, 2)
    > beta.true <- runif(JJ, -1, 1)
    > tau1 <- runif(JJ, -1.5, -0.5)
    > tau2 <- runif(JJ, 0.5, 1.5)
    > ystar <- x.true %o% beta.true + rnorm(NN *JJ)
    >
    > ## These parameters are not needed, but correspond to reparameterized model
    > #d.true <- tau2 - tau1
    > #dd.true <- d.true^2
    > #tau_star <- -tau1/d.true
    > #beta_star <- beta.true/d.true
    >
    > ## Generate roll call matrix using simulated parameters
    > newrc <- matrix(0, NN, JJ)
    > for(j in 1:JJ) newrc[,j] <- cut(ystar[,j], c(-100, tau1[j], tau2[j],100), labels=FALSE)
    >
    > ## Generate starts and priors
    > cur <- vector(mode = "list")
    > cur$DD <- matrix(rep(0.5,JJ), ncol=1)
    > cur$tau <- matrix(rep(-0.5,JJ), ncol=1)
    > cur$beta <- matrix(runif(JJ,-1,1), ncol=1)
    > cur$x <- matrix(runif(NN,-1,1), ncol=1)
    > priors <- vector(mode = "list")
    > priors$x <- list(mu = matrix(0,1,1), sigma = matrix(1,1,1) )
    > priors$beta <- list(mu = matrix(0,2,1), sigma = matrix(diag(25,2),2,2))
    >
    > ## Call ordIRT() with inputs
    > time <- system.time({
    + lout <- ordIRT(.rc = newrc,
    + .starts = cur,
    + .priors = priors,
    + .control = {list(
    + threads = 1,
    + verbose = TRUE,
    + thresh = 1e-6,
    + maxit=300,
    + checkfreq=50
    + )})
    + })
    
    ====================
    ordIRT: Ordinal IRT via Expectation Maximization
    
    Iteration: 50
    
    warning: inv_sympd(): given matrix is not symmetric
    
    warning: inv_sympd(): given matrix is not symmetric
    
    error: inv_sympd(): matrix is singular or not positive definite
    error:
    inv_sympd(): matrix is singular or not positive definite
    terminate called after throwing an instance of 'std::runtime_error'
     what(): inv_sympd(): matrix is singular or not positive definite
    terminate called recursively
Flavor: r-patched-solaris-x86

Version: 0.0.8
Check: whether package can be installed
Result: ERROR
    Installation failed.
Flavors: r-release-osx-x86_64, r-oldrel-osx-x86_64