CRAN Package Check Results for Package gamlss.nl

Last updated on 2019-08-19 07:49:36 CEST.

Flavor Version Tinstall Tcheck Ttotal Status Flags
r-devel-linux-x86_64-debian-clang 4.1-0 7.75 64.28 72.03 ERROR
r-devel-linux-x86_64-debian-gcc 4.1-0 6.97 53.02 59.99 NOTE
r-devel-linux-x86_64-fedora-clang 4.1-0 88.99 NOTE
r-devel-linux-x86_64-fedora-gcc 4.1-0 89.75 NOTE
r-devel-windows-ix86+x86_64 4.1-0 19.00 104.00 123.00 NOTE
r-patched-linux-x86_64 4.1-0 7.33 62.80 70.13 ERROR
r-patched-solaris-x86 4.1-0 137.50 NOTE
r-release-linux-x86_64 4.1-0 6.22 62.30 68.52 ERROR
r-release-windows-ix86+x86_64 4.1-0 14.00 82.00 96.00 NOTE
r-release-osx-x86_64 4.1-0 NOTE
r-oldrel-windows-ix86+x86_64 4.1-0 14.00 109.00 123.00 NOTE
r-oldrel-osx-x86_64 4.1-0 NOTE

Check Details

Version: 4.1-0
Check: dependencies in R code
Result: NOTE
    Package in Depends field not imported from: 'gamlss'
     These packages need to be imported from (in the NAMESPACE file)
     for when this namespace is loaded but not attached.
    ':::' call which should be '::': 'gamlss:::.gamlss.bi.list'
     See the note in ?`:::` about the use of this operator.
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-windows-ix86+x86_64, r-patched-linux-x86_64, r-patched-solaris-x86, r-release-linux-x86_64, r-release-windows-ix86+x86_64, r-release-osx-x86_64, r-oldrel-windows-ix86+x86_64, r-oldrel-osx-x86_64

Version: 4.1-0
Check: R code for possible problems
Result: NOTE
    finterp.data.frame: no visible global function definition for
     'as.formula'
    finterp.data.frame: no visible global function definition for 'terms'
    finterp.data.frame: no visible global function definition for
     'model.matrix'
    finterp.data.frame: no visible global function definition for
     'model.frame'
    finterp.default: no visible global function definition for 'terms'
    finterp.default: no visible global function definition for
     'model.matrix'
    finterp.default: no visible global function definition for
     'model.frame'
    gamlss.nl: no visible global function definition for 'nlm'
    nl: no visible global function definition for 'is'
    nlgamlss: no visible global function definition for 'NO'
    nlgamlss: no visible global function definition for 'as.gamlss.family'
    nlgamlss: no visible global function definition for 'nlm'
    nlgamlss: no visible global function definition for 'cov2cor'
    summary.nlgamlss: no visible global function definition for 'coef'
    summary.nlgamlss: no visible global function definition for 'pnorm'
    Undefined global functions or variables:
     NO as.formula as.gamlss.family coef cov2cor is model.frame
     model.matrix nlm pnorm terms
    Consider adding
     importFrom("methods", "is")
     importFrom("stats", "as.formula", "coef", "cov2cor", "model.frame",
     "model.matrix", "nlm", "pnorm", "terms")
    to your NAMESPACE file (and ensure that your DESCRIPTION Imports field
    contains 'methods').
    
    Found the following calls to attach():
    File 'gamlss.nl/R/nlgamlss.R':
     attach(data)
    See section 'Good practice' in '?attach'.
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: 4.1-0
Check: examples
Result: ERROR
    Running examples in 'gamlss.nl-Ex.R' failed
    The error most likely occurred in:
    
    > base::assign(".ptime", proc.time(), pos = "CheckExEnv")
    > ### Name: gamlss-nl-package
    > ### Title: The GAMLSS add on package for fiting parametric non linear
    > ### models
    > ### Aliases: gamlss-nl-package gamlss-nl
    > ### Keywords: package
    >
    > ### ** Examples
    >
    > data(la)
    > # fitting the Johnson's Su distribtion to the data
    > modJSU <- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
    + nu.fo=~1, mu.start = c(.6, 110), sigma.start= 3, nu.start=1,
    + tau.start=0.6, family=JSU, data=la)
    Error in cov2cor(cov) : 'V' is not a square numeric matrix
    Calls: nlgamlss -> cov2cor
    Execution halted
Flavors: r-devel-linux-x86_64-debian-clang, r-patched-linux-x86_64, r-release-linux-x86_64

Version: 4.1-0
Check: tests
Result: ERROR
     Running 'tests-nlgamlss.R' [4s/5s]
    Running the tests in 'tests/tests-nlgamlss.R' failed.
    Complete output:
     > #------------------------------------------------------------------------------------------
     > # bring the lange data
     > #
     > library(gamlss.nl)
     Loading required package: gamlss
     Loading required package: splines
     Loading required package: gamlss.data
    
     Attaching package: 'gamlss.data'
    
     The following object is masked from 'package:datasets':
    
     sleep
    
     Loading required package: gamlss.dist
     Loading required package: MASS
     Loading required package: nlme
     Loading required package: parallel
     ********** GAMLSS Version 5.1-4 **********
     For more on GAMLSS look at http://www.gamlss.org/
     Type gamlssNews() to see new features/changes/bug fixes.
    
     Loading required package: survival
     > # bring the lange data
     > data(la)
     > plot(PET60~bflow,data=la)
     >
     > #source("C:/GAMLSS/New Functions/logNO.R")
     >
     > #----------------------------------------------------------------------------------------
     > # fiiting the log Normal
     >
     > modLOGNO<- nlgamlss(y=PET60, mu.fo=~log(bflow)+log(1-(1-exp(p1))*exp(-p2/bflow)),
     + sigma.formula=~1, mu.start = c(-.99, 110), sigma.start= -2,
     + family=LOGNO, data=la)
     Warning message:
     In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modLOGNO)-2293.9) > 0.1) stop("error in nl gamlss log-NO")
     > #-----------------------------------
     > # fitting the Normal
     > #with original parameterization
     > modNO<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + mu.start = c(.6, 90), sigma.start= 3, data=la)
     > if(abs(deviance(modNO)-2278.7) > 0.1) stop("error in nl gamlss NO")
     > # with different parameterizaion
     > modNO1<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-(1-exp(p1))*exp(-p2/bflow)),
     + sigma.formula=~1, mu.start = c(-.9, 90), sigma.start= 3, data=la)
     > if(abs(deviance(modNO1)-2278.68) > 0.1) stop("error in nl gamlss NO")
     > # as an example of using function
     > funnl<- function(p) bflow*(1-p[1]*exp(-p[2]/bflow))
     > modNO2<- nlgamlss(y=PET60, mu.fo= funnl, sigma.formula=~1,
     + mu.start = c(.6, 90), sigma.start= 3, data=la)
     > if(abs(deviance(modNO2)-2278.68) > 0.1) stop("error in nl gamlss NO")
     > #-----------------------------------
     > # fitting the Gumbel
     > modGU<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + mu.start = c(.6, 110), sigma.start= 3, family=GU, data=la)
     > if(abs(deviance(modGU)-2382.9) > 0.1) stop("error in nl gamlss GU")
     > modGU1<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-(1-exp(p1))*exp(-p2/bflow)),
     + sigma.formula=~1, mu.start = c(-.6, 90), sigma.start= 3, family=GU, data=la)
     > if(abs(deviance(modGU1)-2382.9) > 0.1) stop("error in nl gamlss GU")
     > #-----------------------------------
     > # fitting the reverse Gumber
     > modRG<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + mu.start = c(.6, 110), sigma.start= 3, family=RG, data=la)
     > if(abs(deviance(modRG)-2249.82) > 0.1) stop("error in nl gamlss RG")
     > #-----------------------------------
     > # fitting Gamma
     > modGA<- nlgamlss(y=PET60, mu.fo= ~log(bflow)+log(1-(1-exp(p1))*exp(-p2/bflow)),
     + sigma.formula=~1, mu.start = c(-.99, 90),, sigma.start= -.5, family=GA, data=la)
     Warning messages:
     1: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     2: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modGA)-2299.87) > 0.1) stop("error in nl gamlss GA")
     > #-----------------------------------
     > # fitting Inverse Gaussian
     > modIG<- nlgamlss(y=PET60, mu.fo= ~log(bflow)+log(1-(1-exp(p1))*exp(-p2/bflow)),
     + sigma.formula=~1, mu.start = c(-.99, 90),, sigma.start= -.5, family=IG, data=la)
     Warning messages:
     1: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     2: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     3: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modIG)-2408.33) > 0.1) stop("error in nl gamlss IG")
     > #-----------------------------------
     > # getting the AIC
     > AIC(modLOGNO,modNO, modGU, modRG, modGA, modIG, k=0)
     df AIC
     modRG 3 2249.818
     modNO 3 2278.682
     modLOGNO 3 2293.899
     modGA 3 2299.873
     modGU 3 2382.901
     modIG 3 2408.328
     > #----------------------------------------------------------------------------------------
     > # three parameters distributions
     > modTF<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1, nu.fo=~1,
     + mu.start = c(.6, 110), sigma.start= 3, nu.start=2.5 ,family=TF, data=la)
     > if(abs(deviance(modTF)-2273.5) > 0.1) stop("error in nl gamlss TF")
     > #-----------------------------------
     > modPE<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + nu.fo=~1, mu.start = c(.6, 110), sigma.start= 3, nu.start=0.6 ,family=PE, data=la)
     Warning message:
     In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modPE)-2275.89) > 0.1) stop("error in nl gamlss PE")
     > #-----------------------------------
     > modBCCG<- nlgamlss(y=PET60, mu.fo=~bflow*(1-(1-exp(p1))*exp(-p2/bflow)),sigma.formula=~1,
     + nu.fo=~1, mu.start = c(-.9, 90), sigma.start= -2.3, nu.start=0, family=BCCG, data=la)
     Warning messages:
     1: In nlgamlss(y = PET60, mu.fo = ~bflow * (1 - (1 - exp(p1)) * exp(-p2/bflow)), :
     the value of typsize supplied is zero or negative the default value of abs(p0) was used instead
     2: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modBCCG)- 2293.74) > 0.1) stop("error in nl gamlss BCCG")
     > AIC(modTF,modPE, modBCCG,k=0)
     df AIC
     modTF 4 2273.486
     modPE 4 2275.889
     modBCCG 4 2293.739
     > #----------------------------------------------------------------------------------------
     > # four parameters
     > # SEP
     > modSEP<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + nu.fo=~1, mu.start = c(.6, 110), sigma.start= 3, nu.start=1, tau.start=0.6,
     + family=SEP, data=la)
     Warning messages:
     1: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     2: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modSEP)-2273.75) > 0.1) stop("error in nl gamlss SEP")
     > #------------------------------------
     > # the BCT
     > modBCT<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-(1-exp(p1))*exp(-p2/bflow)), sigma.formula=~1,
     + nu.fo=~1, mu.start=c(-.9, 90), sigma.start= -2.3, nu.start=0, tau.start=log(2.5),
     + family=BCT, data=la, control=NL.control(hessian=FALSE))
     Warning message:
     In nlgamlss(y = PET60, mu.fo = ~bflow * (1 - (1 - exp(p1)) * exp(-p2/bflow)), :
     the value of typsize supplied is zero or negative the default value of abs(p0) was used instead
     > if((deviance(modBCT)-2293.74) > 0.1) stop("error in nl gamlss BCT")
     > #------------------------------------
     > # BCPE
     > modBCPE<- nlgamlss(y=PET60, mu.fo=~bflow*(1-(1-exp(p1))*exp(-p2/bflow)),sigma.formula=~1,
     + mu.start = c(-.9, 90), sigma.start= -2.3, nu.start=0, tau.start=log(2.5),
     + family=BCPE, data=la)
     Warning messages:
     1: In nlgamlss(y = PET60, mu.fo = ~bflow * (1 - (1 - exp(p1)) * exp(-p2/bflow)), :
     the value of typsize supplied is zero or negative the default value of abs(p0) was used instead
     2: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     3: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if((deviance(modBCPE)- 2292.81) > 0.1) stop("error in nl gamlss BCPE")
     > #------------------------------------
     > # ST3
     > #source("C:/GAMLSS/Distributions_26_11_04/Continuous/ST3/ST3.R")
     > #modST3<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     > #nu.fo=~1, mu.start = c(.6, 110), sigma.start= 3, nu.start=1, tau.start=0.6,
     > #family=ST3, data=la)
     > #if(abs(deviance(modST3)-2266.67) > 0.1) stop("error in nl gamlss ST3")
     > #------------------------------------
     > # JSU
     > modJSU<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + nu.fo=~1, mu.start = c(.6, 110), sigma.start= 3, nu.start=1, tau.start=0.6,
     + family=JSU, data=la)
     Error in cov2cor(cov) : 'V' is not a square numeric matrix
     Calls: nlgamlss -> cov2cor
     Execution halted
Flavors: r-devel-linux-x86_64-debian-clang, r-release-linux-x86_64

Version: 4.1-0
Check: R code for possible problems
Result: NOTE
    finterp.data.frame: no visible global function definition for
     ‘as.formula’
    finterp.data.frame: no visible global function definition for ‘terms’
    finterp.data.frame: no visible global function definition for
     ‘model.matrix’
    finterp.data.frame: no visible global function definition for
     ‘model.frame’
    finterp.default: no visible global function definition for ‘terms’
    finterp.default: no visible global function definition for
     ‘model.matrix’
    finterp.default: no visible global function definition for
     ‘model.frame’
    gamlss.nl: no visible global function definition for ‘nlm’
    nl: no visible global function definition for ‘is’
    nlgamlss: no visible global function definition for ‘NO’
    nlgamlss: no visible global function definition for ‘as.gamlss.family’
    nlgamlss: no visible global function definition for ‘nlm’
    nlgamlss: no visible global function definition for ‘cov2cor’
    summary.nlgamlss: no visible global function definition for ‘coef’
    summary.nlgamlss: no visible global function definition for ‘pnorm’
    Undefined global functions or variables:
     NO as.formula as.gamlss.family coef cov2cor is model.frame
     model.matrix nlm pnorm terms
    Consider adding
     importFrom("methods", "is")
     importFrom("stats", "as.formula", "coef", "cov2cor", "model.frame",
     "model.matrix", "nlm", "pnorm", "terms")
    to your NAMESPACE file (and ensure that your DESCRIPTION Imports field
    contains 'methods').
Flavors: r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc, r-devel-windows-ix86+x86_64, r-patched-solaris-x86, r-release-windows-ix86+x86_64, r-release-osx-x86_64, r-oldrel-windows-ix86+x86_64, r-oldrel-osx-x86_64

Version: 4.1-0
Check: tests
Result: ERROR
     Running ‘tests-nlgamlss.R’ [4s/4s]
    Running the tests in ‘tests/tests-nlgamlss.R’ failed.
    Complete output:
     > #------------------------------------------------------------------------------------------
     > # bring the lange data
     > #
     > library(gamlss.nl)
     Loading required package: gamlss
     Loading required package: splines
     Loading required package: gamlss.data
    
     Attaching package: 'gamlss.data'
    
     The following object is masked from 'package:datasets':
    
     sleep
    
     Loading required package: gamlss.dist
     Loading required package: MASS
     Loading required package: nlme
     Loading required package: parallel
     ********** GAMLSS Version 5.1-4 **********
     For more on GAMLSS look at http://www.gamlss.org/
     Type gamlssNews() to see new features/changes/bug fixes.
    
     Loading required package: survival
     > # bring the lange data
     > data(la)
     > plot(PET60~bflow,data=la)
     >
     > #source("C:/GAMLSS/New Functions/logNO.R")
     >
     > #----------------------------------------------------------------------------------------
     > # fiiting the log Normal
     >
     > modLOGNO<- nlgamlss(y=PET60, mu.fo=~log(bflow)+log(1-(1-exp(p1))*exp(-p2/bflow)),
     + sigma.formula=~1, mu.start = c(-.99, 110), sigma.start= -2,
     + family=LOGNO, data=la)
     Warning message:
     In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modLOGNO)-2293.9) > 0.1) stop("error in nl gamlss log-NO")
     > #-----------------------------------
     > # fitting the Normal
     > #with original parameterization
     > modNO<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + mu.start = c(.6, 90), sigma.start= 3, data=la)
     > if(abs(deviance(modNO)-2278.7) > 0.1) stop("error in nl gamlss NO")
     > # with different parameterizaion
     > modNO1<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-(1-exp(p1))*exp(-p2/bflow)),
     + sigma.formula=~1, mu.start = c(-.9, 90), sigma.start= 3, data=la)
     > if(abs(deviance(modNO1)-2278.68) > 0.1) stop("error in nl gamlss NO")
     > # as an example of using function
     > funnl<- function(p) bflow*(1-p[1]*exp(-p[2]/bflow))
     > modNO2<- nlgamlss(y=PET60, mu.fo= funnl, sigma.formula=~1,
     + mu.start = c(.6, 90), sigma.start= 3, data=la)
     > if(abs(deviance(modNO2)-2278.68) > 0.1) stop("error in nl gamlss NO")
     > #-----------------------------------
     > # fitting the Gumbel
     > modGU<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + mu.start = c(.6, 110), sigma.start= 3, family=GU, data=la)
     > if(abs(deviance(modGU)-2382.9) > 0.1) stop("error in nl gamlss GU")
     > modGU1<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-(1-exp(p1))*exp(-p2/bflow)),
     + sigma.formula=~1, mu.start = c(-.6, 90), sigma.start= 3, family=GU, data=la)
     > if(abs(deviance(modGU1)-2382.9) > 0.1) stop("error in nl gamlss GU")
     > #-----------------------------------
     > # fitting the reverse Gumber
     > modRG<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + mu.start = c(.6, 110), sigma.start= 3, family=RG, data=la)
     > if(abs(deviance(modRG)-2249.82) > 0.1) stop("error in nl gamlss RG")
     > #-----------------------------------
     > # fitting Gamma
     > modGA<- nlgamlss(y=PET60, mu.fo= ~log(bflow)+log(1-(1-exp(p1))*exp(-p2/bflow)),
     + sigma.formula=~1, mu.start = c(-.99, 90),, sigma.start= -.5, family=GA, data=la)
     Warning messages:
     1: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     2: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modGA)-2299.87) > 0.1) stop("error in nl gamlss GA")
     > #-----------------------------------
     > # fitting Inverse Gaussian
     > modIG<- nlgamlss(y=PET60, mu.fo= ~log(bflow)+log(1-(1-exp(p1))*exp(-p2/bflow)),
     + sigma.formula=~1, mu.start = c(-.99, 90),, sigma.start= -.5, family=IG, data=la)
     Warning messages:
     1: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     2: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     3: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modIG)-2408.33) > 0.1) stop("error in nl gamlss IG")
     > #-----------------------------------
     > # getting the AIC
     > AIC(modLOGNO,modNO, modGU, modRG, modGA, modIG, k=0)
     df AIC
     modRG 3 2249.818
     modNO 3 2278.682
     modLOGNO 3 2293.899
     modGA 3 2299.873
     modGU 3 2382.901
     modIG 3 2408.328
     > #----------------------------------------------------------------------------------------
     > # three parameters distributions
     > modTF<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1, nu.fo=~1,
     + mu.start = c(.6, 110), sigma.start= 3, nu.start=2.5 ,family=TF, data=la)
     > if(abs(deviance(modTF)-2273.5) > 0.1) stop("error in nl gamlss TF")
     > #-----------------------------------
     > modPE<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + nu.fo=~1, mu.start = c(.6, 110), sigma.start= 3, nu.start=0.6 ,family=PE, data=la)
     Warning message:
     In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modPE)-2275.89) > 0.1) stop("error in nl gamlss PE")
     > #-----------------------------------
     > modBCCG<- nlgamlss(y=PET60, mu.fo=~bflow*(1-(1-exp(p1))*exp(-p2/bflow)),sigma.formula=~1,
     + nu.fo=~1, mu.start = c(-.9, 90), sigma.start= -2.3, nu.start=0, family=BCCG, data=la)
     Warning messages:
     1: In nlgamlss(y = PET60, mu.fo = ~bflow * (1 - (1 - exp(p1)) * exp(-p2/bflow)), :
     the value of typsize supplied is zero or negative the default value of abs(p0) was used instead
     2: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modBCCG)- 2293.74) > 0.1) stop("error in nl gamlss BCCG")
     > AIC(modTF,modPE, modBCCG,k=0)
     df AIC
     modTF 4 2273.486
     modPE 4 2275.889
     modBCCG 4 2293.739
     > #----------------------------------------------------------------------------------------
     > # four parameters
     > # SEP
     > modSEP<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + nu.fo=~1, mu.start = c(.6, 110), sigma.start= 3, nu.start=1, tau.start=0.6,
     + family=SEP, data=la)
     Warning messages:
     1: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     2: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if(abs(deviance(modSEP)-2273.75) > 0.1) stop("error in nl gamlss SEP")
     > #------------------------------------
     > # the BCT
     > modBCT<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-(1-exp(p1))*exp(-p2/bflow)), sigma.formula=~1,
     + nu.fo=~1, mu.start=c(-.9, 90), sigma.start= -2.3, nu.start=0, tau.start=log(2.5),
     + family=BCT, data=la, control=NL.control(hessian=FALSE))
     Warning message:
     In nlgamlss(y = PET60, mu.fo = ~bflow * (1 - (1 - exp(p1)) * exp(-p2/bflow)), :
     the value of typsize supplied is zero or negative the default value of abs(p0) was used instead
     > if((deviance(modBCT)-2293.74) > 0.1) stop("error in nl gamlss BCT")
     > #------------------------------------
     > # BCPE
     > modBCPE<- nlgamlss(y=PET60, mu.fo=~bflow*(1-(1-exp(p1))*exp(-p2/bflow)),sigma.formula=~1,
     + mu.start = c(-.9, 90), sigma.start= -2.3, nu.start=0, tau.start=log(2.5),
     + family=BCPE, data=la)
     Warning messages:
     1: In nlgamlss(y = PET60, mu.fo = ~bflow * (1 - (1 - exp(p1)) * exp(-p2/bflow)), :
     the value of typsize supplied is zero or negative the default value of abs(p0) was used instead
     2: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     3: In nlm(optFunction, p = p0, hessian = hessian, fscale = fscale, :
     NA/Inf replaced by maximum positive value
     > if((deviance(modBCPE)- 2292.81) > 0.1) stop("error in nl gamlss BCPE")
     > #------------------------------------
     > # ST3
     > #source("C:/GAMLSS/Distributions_26_11_04/Continuous/ST3/ST3.R")
     > #modST3<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     > #nu.fo=~1, mu.start = c(.6, 110), sigma.start= 3, nu.start=1, tau.start=0.6,
     > #family=ST3, data=la)
     > #if(abs(deviance(modST3)-2266.67) > 0.1) stop("error in nl gamlss ST3")
     > #------------------------------------
     > # JSU
     > modJSU<- nlgamlss(y=PET60, mu.fo= ~bflow*(1-p1*exp(-p2/bflow)), sigma.formula=~1,
     + nu.fo=~1, mu.start = c(.6, 110), sigma.start= 3, nu.start=1, tau.start=0.6,
     + family=JSU, data=la)
     Error in cov2cor(cov) : 'V' is not a square numeric matrix
     Calls: nlgamlss -> cov2cor
     Execution halted
Flavor: r-patched-linux-x86_64