CRAN Package Check Results for Package restriktor

Last updated on 2018-04-27 05:50:14 CEST.

Flavor Version Tinstall Tcheck Ttotal Status Flags
r-devel-linux-x86_64-debian-clang 0.1-80.711 9.37 57.32 66.69 ERROR
r-devel-linux-x86_64-debian-gcc 0.1-80.711 8.81 51.04 59.85 ERROR
r-devel-linux-x86_64-fedora-clang 0.1-80.711 97.34 OK
r-devel-linux-x86_64-fedora-gcc 0.1-80.711 92.02 OK
r-devel-windows-ix86+x86_64 0.1-80.711 19.00 83.00 102.00 OK
r-patched-linux-x86_64 0.1-80.711 11.77 64.89 76.66 ERROR
r-patched-solaris-x86 0.1-80.711 133.60 OK
r-release-linux-x86_64 0.1-80.711 9.61 64.91 74.52 ERROR
r-release-windows-ix86+x86_64 0.1-80.711 19.00 83.00 102.00 OK
r-release-osx-x86_64 0.1-80.711 OK
r-oldrel-windows-ix86+x86_64 0.1-80.711 7.00 120.00 127.00 OK
r-oldrel-osx-x86_64 0.1-80.711 OK

Check Details

Version: 0.1-80.711
Check: examples
Result: ERROR
    Running examples in ‘restriktor-Ex.R’ failed
    The error most likely occurred in:
    
    > base::assign(".ptime", proc.time(), pos = "CheckExEnv")
    > ### Name: conTest
    > ### Title: function for informative hypothesis testing (iht)
    > ### Aliases: conTest iht
    >
    > ### ** Examples
    >
    > ## example 1:
    > # the data consist of ages (in months) at which an
    > # infant starts to walk alone.
    >
    > # prepare data
    > DATA1 <- subset(ZelazoKolb1972, Group != "Control")
    >
    > # fit unrestricted linear model
    > fit1.lm <- lm(Age ~ -1 + Group, data = DATA1)
    >
    > # the variable names can be used to impose constraints on
    > # the corresponding regression parameters.
    > coef(fit1.lm)
     GroupActive GroupNo GroupPassive
     10.125 12.350 11.375
    >
    > # constraint syntax: assuming that the walking
    > # exercises would not have a negative effect of increasing the
    > # mean age at which a child starts to walk.
    > myConstraints1 <- ' GroupActive < GroupPassive;
    + GroupPassive < GroupNo '
    >
    > conTest(fit1.lm, myConstraints1)
    
    Restriktor: restricted hypothesis tests ( 14 residual degrees of freedom ):
    
    
    Multiple R-squared remains 0.985
    
    Constraint matrix:
     GroupActive GroupNo GroupPassive op rhs active
    1: -1 0 1 >= 0 no
    2: 0 1 -1 >= 0 no
    
    
    Overview of all available hypothesis tests:
    
    Global test: H0: all parameters are restricted to be equal (==)
     vs. HA: at least one inequality restriction is strictly true (>)
     Test statistic: 5.9780, p-value: 0.02834
    
    Type A test: H0: all restrictions are equalities (==)
     vs. HA: at least one inequality restriction is strictly true (>)
     Test statistic: 5.9780, p-value: 0.02834
    
    Type B test: H0: all restrictions hold in the population
     vs. HA: at least one restriction is violated
     Test statistic: 0.0000, p-value: 1
    
    Type C test: H0: at least one restriction is false or active (==)
     vs. HA: all restrictions are strictly true (>)
     Test statistic: 1.0627, p-value: 0.1529
    
    Note: Type C test is based on a t-distribution (one-sided),
     all other tests are based on a mixture of F-distributions.
    
    >
    >
    > # another way is to first fit the restricted model
    > fit.restr1 <- restriktor(fit1.lm, constraints = myConstraints1)
    Error in UseMethod("conLM") :
     no applicable method for 'conLM' applied to an object of class "lm"
    Calls: restriktor
    Execution halted
Flavors: r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc, r-patched-linux-x86_64, r-release-linux-x86_64

Version: 0.1-80.711
Check: tests
Result: ERROR
     Running ‘testCases.R’ [1s/1s]
    Running the tests in ‘tests/testCases.R’ failed.
    Complete output:
     > ## test constrained lm, rlm, glm
     > # mixing weights
     > # bootstrap standard errors
     > # models with and without intercept
     > # case weights
     > # constraints on factors and interactions
     >
     > library(restriktor)
     This is restriktor 0.1-80.711
     restriktor is BETA software! Please report any bugs.
     > n <- 100
     > p <- 4
     > betas = c(0.1,0.2,0.3,0.4,0.5)
     > set.seed(3013073)
     > X <- cbind(mvtnorm:::rmvnorm(n, mean=rep(0,p), sigma=diag(p)), rbinom(n,1,0.5))
     > colnames(X) <- c("x1","x2","x3","x4","f1")
     > z <- X %*% betas
     > y <- z + rnorm(n)
     > DATA <- data.frame(y, X)
     >
     > # intercept model
     > model1 <- y ~ 1 + x1 + x2 + x3 + x4
     > # no intercept model
     > model2 <- y ~ -1 + x1 + x2 + x3 + x4
     > # intercept model with interaction
     > model3 <- y ~ 1 + x1*f1 + x2*f1 + x3*f1 + x4*f1
     > # intercept model with interaction
     > model4 <- y ~ -1 + x1*f1 + x2*f1 + x3*f1 + x4*f1
     >
     > ############################ lm #################################
     > linmod1 <- lm(model1, data = DATA)
     > linmod2 <- lm(model2, data = DATA)
     > linmod3 <- lm(model3, data = DATA)
     > linmod1wt <- lm(model1, data = DATA, weights = abs(rnorm(n)))
     > linmod2wt <- lm(model2, data = DATA, weights = abs(rnorm(n)))
     > linmod1fac <- lm(model3, data = DATA)
     > linmod2fac <- lm(model4, data = DATA)
     >
     > constraints1 <- 'x1 > 0; x2 > 0; x3 > 0'
     > fit.restr1 <- restriktor(linmod1, constraints1)
     Error in UseMethod("conLM") :
     no applicable method for 'conLM' applied to an object of class "lm"
     Calls: restriktor
     Execution halted
Flavors: r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc

Version: 0.1-80.711
Check: tests
Result: ERROR
     Running ‘testCases.R’ [1s/2s]
    Running the tests in ‘tests/testCases.R’ failed.
    Complete output:
     > ## test constrained lm, rlm, glm
     > # mixing weights
     > # bootstrap standard errors
     > # models with and without intercept
     > # case weights
     > # constraints on factors and interactions
     >
     > library(restriktor)
     This is restriktor 0.1-80.711
     restriktor is BETA software! Please report any bugs.
     > n <- 100
     > p <- 4
     > betas = c(0.1,0.2,0.3,0.4,0.5)
     > set.seed(3013073)
     > X <- cbind(mvtnorm:::rmvnorm(n, mean=rep(0,p), sigma=diag(p)), rbinom(n,1,0.5))
     > colnames(X) <- c("x1","x2","x3","x4","f1")
     > z <- X %*% betas
     > y <- z + rnorm(n)
     > DATA <- data.frame(y, X)
     >
     > # intercept model
     > model1 <- y ~ 1 + x1 + x2 + x3 + x4
     > # no intercept model
     > model2 <- y ~ -1 + x1 + x2 + x3 + x4
     > # intercept model with interaction
     > model3 <- y ~ 1 + x1*f1 + x2*f1 + x3*f1 + x4*f1
     > # intercept model with interaction
     > model4 <- y ~ -1 + x1*f1 + x2*f1 + x3*f1 + x4*f1
     >
     > ############################ lm #################################
     > linmod1 <- lm(model1, data = DATA)
     > linmod2 <- lm(model2, data = DATA)
     > linmod3 <- lm(model3, data = DATA)
     > linmod1wt <- lm(model1, data = DATA, weights = abs(rnorm(n)))
     > linmod2wt <- lm(model2, data = DATA, weights = abs(rnorm(n)))
     > linmod1fac <- lm(model3, data = DATA)
     > linmod2fac <- lm(model4, data = DATA)
     >
     > constraints1 <- 'x1 > 0; x2 > 0; x3 > 0'
     > fit.restr1 <- restriktor(linmod1, constraints1)
     Error in UseMethod("conLM") :
     no applicable method for 'conLM' applied to an object of class "lm"
     Calls: restriktor
     Execution halted
Flavors: r-patched-linux-x86_64, r-release-linux-x86_64