CRAN Package Check Results for Package LambertW

Last updated on 2014-10-25 19:47:39.

Flavor Version Tinstall Tcheck Ttotal Status Flags
r-devel-linux-x86_64-debian-clang 0.2.9.9.5 1.47 32.71 34.18 NOTE
r-devel-linux-x86_64-debian-gcc 0.2.9.9.5 1.43 32.88 34.31 NOTE
r-devel-linux-x86_64-fedora-clang 0.2.9.9.5 66.72 NOTE
r-devel-linux-x86_64-fedora-gcc 0.2.9.9.5 63.30 NOTE
r-devel-osx-x86_64-clang 0.2.9.9.5 60.87 NOTE
r-devel-windows-ix86+x86_64 0.2.9.9.5 6.00 51.00 57.00 NOTE
r-patched-linux-x86_64 0.2.9.9.5 1.52 33.23 34.74 NOTE
r-patched-solaris-sparc 0.2.9.9.5 382.00 NOTE
r-patched-solaris-x86 0.2.9.9.5 84.30 NOTE
r-release-linux-ix86 0.2.9.9.5 1.96 45.24 47.20 OK
r-release-linux-x86_64 0.2.9.9.5 1.45 34.00 35.44 OK
r-release-osx-x86_64-mavericks 0.2.9.9.5 ERROR
r-release-osx-x86_64-snowleopard 0.2.9.9.5 OK
r-release-windows-ix86+x86_64 0.2.9.9.5 7.00 38.00 45.00 ERROR
r-oldrel-windows-ix86+x86_64 0.2.9.9.5 6.00 56.00 62.00 OK

Check Details

Version: 0.2.9.9.5
Check: R code for possible problems
Result: NOTE
    MLE_LambertW_new: no visible global function definition for ‘solnp’
Flavors: r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc, r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc, r-devel-osx-x86_64-clang, r-devel-windows-ix86+x86_64, r-patched-linux-x86_64, r-patched-solaris-sparc, r-patched-solaris-x86

Version: 0.2.9.9.5
Check: package dependencies
Result: ERROR
    Package required but not available: ‘gsl’
    
    See the information on DESCRIPTION files in the chapter ‘Creating R
    packages’ of the ‘Writing R Extensions’ manual.
Flavor: r-release-osx-x86_64-mavericks

Version: 0.2.9.9.5
Check: examples
Result: ERROR
    Running examples in 'LambertW-Ex.R' failed
    The error most likely occurred in:
    
    > ### Name: LambertW-package
    > ### Title: Lambert W Random Variables
    > ### Aliases: LambertW-package LambertW
    > ### Keywords: package
    >
    > ### ** Examples
    >
    > ## Replicate parts of the statistical analysis in the References (2011a)
    > data(AA)
    > attach(AA)
    > X=AA[AA$sex=="f",]
    > y=X$bmi
    >
    > op=par(no.readonly=TRUE)
    > normfit(y)
    $sw
    
     Shapiro-Wilk normality test
    
    data: data
    W = 0.9725, p-value = 0.03454
    
    
    $sf
    
     Shapiro-Francia normality test
    
    data: data
    W = 0.9699, p-value = 0.02306
    
    
    $ad
    
     Anderson-Darling normality test
    
    data: data
    A = 0.4948, p-value = 0.2104
    
    
    >
    > fit.gmm=IGMM(y, type="s")
    > summary(fit.gmm) # gamma is significant and positive
    Call: IGMM.default(y = y, type = "s")
    Estimation method: IGMM
    Input distribution: Any distribution with theoretical skewness = 0 .
    
     Parameter estimates:
    WARNING: Standard Errors are only asymptotic simulation based!!!
     Estimate Std. Error t value Pr(>|t|)
    mu_x 21.733826 0.100000 217.3383 < 2e-16 ***
    sigma_x 2.569911 0.070711 36.3440 < 2e-16 ***
    gamma 0.099854 0.040000 2.4963 0.01255 *
    ---
    Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
    --------------------------------------------------------------
     a b
    Support 12.26583 Inf
    Data range 16.75000 31.93
    
     p_1 = Probability that non-principal branch affects the solution: NA
    --------------------------------------------------------------
    
    Given these input parameter estimates the corresponding output moments are (assuming Gaussian input):
     mu_y = 21.99; sigma_y = 2.63; skewness = 0.6; kurtosis = 3.62.
    
    > plot(fit.gmm)
    > # Comparison of estimated theoretical and sample moments
    > TAB = rbind(rbind(mLambertW(beta = fit.gmm$tau[1:2],
    + gamma = fit.gmm$tau["gamma"], distname="normal")),
    + cbind(mean(y), sd(y), skewness(y), kurtosis(y)))
    > rownames(TAB) = c("Theoretical (IGMM)", "Sample")
    > TAB
     mean sd skewness kurtosis
    Theoretical (IGMM) 21.99172 2.634325 0.6041175 3.616055
    Sample 21.9892 2.640028 0.6933951 4.176055
    >
    > x=get.input(y, fit.gmm$tau)
    > normfit(x) # input is normal -> fit a Lambert W x Gaussian by maximum likelihood
    $sw
    
     Shapiro-Wilk normality test
    
    data: data
    W = 0.9944, p-value = 0.9571
    
    
    $sf
    
     Shapiro-Francia normality test
    
    data: data
    W = 0.9934, p-value = 0.8417
    
    
    $ad
    
     Anderson-Darling normality test
    
    data: data
    A = 0.1533, p-value = 0.9572
    
    
    >
    > fit.ml=MLE_LambertW(y, type="s", distname="normal")
Flavor: r-release-windows-ix86+x86_64