CRAN Package Check Results for Package sn

Last updated on 2018-01-17 02:53:17 CET.

Flavor Version Tinstall Tcheck Ttotal Status Flags
r-devel-linux-x86_64-debian-clang 1.5-1 1.89 101.50 103.39 OK
r-devel-linux-x86_64-debian-gcc 1.5-1 1.97 72.43 74.40 ERROR
r-devel-linux-x86_64-fedora-clang 1.5-1 117.41 OK
r-devel-linux-x86_64-fedora-gcc 1.5-1 110.60 OK
r-devel-windows-ix86+x86_64 1.5-1 6.00 128.00 134.00 OK
r-patched-linux-x86_64 1.5-1 2.35 108.38 110.73 OK
r-patched-solaris-x86 1.5-1 139.60 OK
r-release-linux-x86_64 1.5-1 2.94 99.47 102.41 OK
r-release-windows-ix86+x86_64 1.5-1 5.00 112.00 117.00 OK
r-release-osx-x86_64 1.5-1 OK
r-oldrel-windows-ix86+x86_64 1.5-1 6.00 90.00 96.00 OK
r-oldrel-osx-x86_64 1.5-1 OK

Check Details

Version: 1.5-1
Check: examples
Result: ERROR
    Running examples in ‘sn-Ex.R’ failed
    The error most likely occurred in:
    
    > base::assign(".ptime", proc.time(), pos = "CheckExEnv")
    > ### Name: selm
    > ### Title: Fitting linear models with skew-elliptical error term
    > ### Aliases: selm
    > ### Keywords: regression univar multivariate
    >
    > ### ** Examples
    >
    > data(ais)
    > m1 <- selm(log(Fe) ~ BMI + LBM, family="SN", data=ais)
    > print(m1)
    Object class: selm
    Call: selm(formula = log(Fe) ~ BMI + LBM, family = "SN", data = ais)
    Number of observations: 202
    Number of covariates: 3 (includes constant term)
    Number of parameters: 5
    Family: SN
    Estimation method: MLE
    Log-likelihood: -175.9094
    > summary(m1)
    Call: selm(formula = log(Fe) ~ BMI + LBM, family = "SN", data = ais)
    Number of observations: 202
    Family: SN
    Estimation method: MLE
    Log-likelihood: -175.9094
    Parameter type: CP
    
    CP residuals:
     Min 1Q Median 3Q Max
    -2.01862 -0.39918 0.03332 0.40744 1.33000
    
    Regression coefficients
     estimate std.err z-ratio Pr{>|z|}
    (Intercept.CP) 2.736823 0.337376 8.112094 0.000
    BMI 0.035270 0.020760 1.698934 0.089
    LBM 0.009465 0.004467 2.118964 0.034
    
    Parameters of the SEC random component
     estimate std.err
    s.d. 0.5796 0.029
    gamma1 -0.1742 0.180
    > s<- summary(m1, "DP", cov=TRUE, cor=TRUE)
    > plot(m1)
    > plot(m1, param.type="DP")
    > logLik(m1)
    'log Lik.' -175.9094 (df=5)
    > coef(m1)
    (Intercept.CP) BMI LBM s.d. gamma1
     2.73682314 0.03526978 0.00946493 0.57960516 -0.17419973
    > coef(m1, "DP")
    (Intercept.DP) BMI LBM omega alpha
     3.16595780 0.03526978 0.00946493 0.72117869 -1.11946607
    > var <- vcov(m1)
    > #
    > m1a <- selm(log(Fe) ~ BMI + LBM, family="SN", method="MPLE", data=ais)
    > m1b <- selm(log(Fe) ~ BMI + LBM, family="ST", fixed.param=list(nu=8), data=ais)
    Warning in st.infoUv(dp = fit$dp, NULL, x, yInfo, w, fixed.nu, symmetr, :
     'dp' does not seem to be at MLE; score not quite 0
    score(dp): -2.188516 -23.17408 -97.11125 82.92764 0.4160464
    norm(score)^2: 8.722453
    Error in chol.default(x, pivot = FALSE) :
     the leading minor of order 5 is not positive definite
    Error in pd.solve(I.cp) : x appears to be not positive definite
    Calls: selm -> selm.fit -> st.infoUv -> pd.solve
    Execution halted
Flavor: r-devel-linux-x86_64-debian-gcc