Last updated on 2026-07-23 00:53:39 CEST.
| Flavor | Version | Tinstall | Tcheck | Ttotal | Status | Flags |
|---|---|---|---|---|---|---|
| r-devel-linux-x86_64-debian-clang | 1.1-8 | 10.67 | 119.52 | 130.19 | NOTE | |
| r-devel-linux-x86_64-debian-gcc | 1.1-8 | 6.66 | 86.40 | 93.06 | ERROR | |
| r-devel-linux-x86_64-fedora-clang | 1.1-8 | 16.00 | 188.87 | 204.87 | OK | |
| r-devel-linux-x86_64-fedora-gcc | 1.1-8 | 87.38 | OK | |||
| r-devel-windows-x86_64 | 1.1-8 | 14.00 | 154.00 | 168.00 | OK | |
| r-patched-linux-x86_64 | 1.1-8 | 10.35 | 118.14 | 128.49 | OK | |
| r-release-linux-x86_64 | 1.1-8 | 8.59 | 115.71 | 124.30 | OK | |
| r-release-macos-arm64 | 1.1-8 | 3.00 | 33.00 | 36.00 | OK | |
| r-release-macos-x86_64 | 1.1-8 | 8.00 | 143.00 | 151.00 | OK | |
| r-release-windows-x86_64 | 1.1-8 | 14.00 | 157.00 | 171.00 | OK | |
| r-oldrel-macos-arm64 | 1.1-8 | OK | ||||
| r-oldrel-macos-x86_64 | 1.1-8 | 7.00 | 117.00 | 124.00 | OK | |
| r-oldrel-windows-x86_64 | 1.1-8 | 17.00 | 190.00 | 207.00 | OK |
Version: 1.1-8
Check: CRAN incoming feasibility
Result: NOTE
Maintainer: ‘Arne Henningsen <arne.henningsen@gmail.com>’
No Authors@R field in DESCRIPTION.
Please add one, modifying
Authors@R: c(person(given = "Tim",
family = "Coelli",
role = "aut"),
person(given = "Arne",
family = "Henningsen",
role = c("aut", "cre"),
email = "arne.henningsen@gmail.com"))
as necessary.
Package CITATION file contains call(s) to old-style personList() or
as.personList(). Please use c() on person objects instead.
Package CITATION file contains call(s) to old-style citEntry(). Please
use bibentry() instead.
Flavors: r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc
Version: 1.1-8
Check: tests
Result: ERROR
Running ‘fail.R’ [1s/1s]
Running ‘frontier41.R’ [1s/1s]
Running ‘frontierTest.R’ [29s/36s]
Running ‘nestedModels.R’ [2s/3s]
Running ‘translogRay.R’ [1s/2s]
Running ‘translogRayMult.R’ [3s/3s]
Running ‘wrongSkewness.R’ [1s/2s]
Running the tests in ‘tests/fail.R’ failed.
Complete output:
> library( "frontier" )
Loading required package: micEcon
If you have questions, suggestions, or comments regarding one of the 'micEcon' packages, please use a forum or 'tracker' at micEcon's R-Forge site:
https://r-forge.r-project.org/projects/micecon/
Loading required package: lmtest
Loading required package: zoo
Attaching package: 'zoo'
The following objects are masked from 'package:base':
as.Date, as.Date.numeric
Please cite the 'frontier' package as:
Tim Coelli and Arne Henningsen (2013). frontier: Stochastic Frontier Analysis. R package version 1.1. http://CRAN.R-Project.org/package=frontier.
If you have questions, suggestions, or comments regarding the 'frontier' package, please use a forum or 'tracker' at frontier's R-Forge site:
https://r-forge.r-project.org/projects/frontier/
> library( "plm" )
> options( digits = 5 )
>
> ## example data included in FRONTIER 4.1 (cross-section data)
> data( front41Data )
> front41Data$firmNo <- c( 1:nrow( front41Data ) )
>
> ## non-existing variable
> try( sfa( log( output ) ~ log( capital7 ) + log( labour ),
+ data = front41Data ) )
Error in eval(predvars, data, env) : object 'capital7' not found
>
> ## nParamTotal > nObs
> try( sfa( log( output ) ~ log( capital ) + log( labour ),
+ data = front41Data[ 1:4, ] ) )
Error in sfa(log(output) ~ log(capital) + log(labour), data = front41Data[1:4, :
the model cannot be estimated, because the number of parameters (5) is larger than the number of observations (4)
>
> ## nParamTotal >> nObs
> try( sfa( log( output ) ~ log( capital ) + log( labour ),
+ data = front41Data[ 1:2, ] ) )
Error in sfa(log(output) ~ log(capital) + log(labour), data = front41Data[1:2, :
the model cannot be estimated, because the number of parameters (5) is larger than the number of observations (2)
>
> ## nParamTotal > number of valid observations
> try( sfa( log( output ) ~ log( capital ) + log( labour ) + log( firmNo - 56 ),
+ data = front41Data ) )
Error in sfa(log(output) ~ log(capital) + log(labour) + log(firmNo - 56), :
the model cannot be estimated, because the number of parameters (6) is larger than the number of valid observations (4)
In addition: Warning message:
In log(firmNo - 56) : NaNs produced
>
> ## the dependent variable has only infinite values
> try( sfa( log( 0 * output ) ~ log( capital ) + log( labour ),
+ data = front41Data ) )
Error in sfa(log(0 * output) ~ log(capital) + log(labour), data = front41Data) :
the dependent variable has no valid observations
>
> ## the dependent variable has only NA values
> try( sfa( log( -output ) ~ log( capital ) + log( labour ),
+ data = front41Data ) )
Error in sfa(log(-output) ~ log(capital) + log(labour), data = front41Data) :
the dependent variable has no valid observations
In addition: Warning message:
In log(-output) : NaNs produced
>
> ## one of the regressors has only infinite values
> try( sfa( log( output ) ~ log( 0 * capital ) + log( labour ),
+ data = front41Data ) )
Error in sfa(log(output) ~ log(0 * capital) + log(labour), data = front41Data) :
regressor 'log(0 * capital)' has no valid observations
>
> ## one of the regressors has only NA values
> try( sfa( log( output ) ~ log( capital ) + log( -labour ),
+ data = front41Data ) )
Error in sfa(log(output) ~ log(capital) + log(-labour), data = front41Data) :
regressor 'log(-labour)' has no valid observations
In addition: Warning message:
In log(-labour) : NaNs produced
>
> ## one of the regressors of the inefficiency term has only infinite values
> try( sfa( log( output ) ~ log( capital ) + log( labour ) | log( 0 * firmNo ),
+ data = front41Data ) )
Error in sfa(log(output) ~ log(capital) + log(labour) | log(0 * firmNo), :
the regressor for the inefficiency term 'log(0 * firmNo)' has no valid observations
>
> ## one of the regressors of the inefficiency term has only NA values
> try( sfa( log( output ) ~ log( capital ) + log( labour ) | log(-firmNo ),
+ data = front41Data ) )
Error in sfa(log(output) ~ log(capital) + log(labour) | log(-firmNo), :
the regressor for the inefficiency term 'log(-firmNo)' has no valid observations
In addition: Warning message:
In log(-firmNo) : NaNs produced
>
> ## no convergence
> a1 <- sfa( log( output ) ~ log( capital ) + log( labour ),
+ data = front41Data, maxit = 2 )
Warning message:
In sfa(log(output) ~ log(capital) + log(labour), data = front41Data, :
the maximum number of iterations has been reached; please try again using different starting values and/or increase the maximum number of iterations
> print( summary( a1 ), digits = 2 )
Error Components Frontier (see Battese & Coelli 1992)
Inefficiency decreases the endogenous variable (as in a production function)
The dependent variable is logged
Iterative ML estimation terminated after 2 iterations:
maximum number of iterations reached
final maximum likelihood estimates
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.58 0.97 0.6 0.6
log(capital) 0.28 0.95 0.3 0.8
log(labour) 0.53 0.40 1.3 0.2
sigmaSq 0.22 1.00 0.2 0.8
gamma 0.80 1.00 0.8 0.4
log likelihood value: -17.033
cross-sectional data
total number of observations = 60
mean efficiency: 0.73919
>
> ## no convergence, L(MLE) < L(OLS)
> a2 <- sfa( log( output ) ~ log( capital ) + log( labour ),
+ data = front41Data, maxit = 2, start = c( 1, 0, 0, 1, 0.5 ) )
Warning message:
In sfa(log(output) ~ log(capital) + log(labour), data = front41Data, :
the maximum number of iterations has been reached and the likelihood value of the ML estimation is less than that obtained using OLS; please try again using different starting values and/or increase the maximum number of iterations
> print( summary( a2 ), digits = 2 )
Error Components Frontier (see Battese & Coelli 1992)
Inefficiency decreases the endogenous variable (as in a production function)
The dependent variable is logged
Iterative ML estimation terminated after 2 iterations:
maximum number of iterations reached
final maximum likelihood estimates
Estimate Std. Error z value Pr(>|z|)
(Intercept) 1.12 0.98 1.1 0.3
log(capital) 0.20 0.96 0.2 0.8
log(labour) 0.51 0.58 0.9 0.4
sigmaSq 0.98 0.98 1.0 0.3
gamma 0.36 0.90 0.4 0.7
log likelihood value: -56.035
cross-sectional data
total number of observations = 60
mean efficiency: 0.64843
>
> ## no convergence, L(MLE) < L(OLS), wrong skewness
> a3 <- sfa( log( output ) ~ log( capital ) + log( labour ),
+ data = front41Data, maxit = 2, ineffDecrease = FALSE )
Warning messages:
1: In sfa(log(output) ~ log(capital) + log(labour), data = front41Data, :
the residuals of the OLS estimates are left-skewed and the likelihood value of the ML estimation is less than that obtained using OLS; this usually indicates that there is no inefficiency or that the model is misspecified
2: In sfa(log(output) ~ log(capital) + log(labour), data = front41Data, :
the maximum number of iterations has been reached; please try again using different starting values and/or increase the maximum number of iterations
> print( summary( a3, effMinusU = FALSE ), digits = 1 )
Error Components Frontier (see Battese & Coelli 1992)
Inefficiency increases the endogenous variable (as in a cost function)
The dependent variable is logged
Iterative ML estimation terminated after 2 iterations:
maximum number of iterations reached
final maximum likelihood estimates
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.19 0.99 0.2 0.9
log(capital) 0.28 0.98 0.3 0.8
log(labour) 0.53 0.77 0.7 0.5
sigmaSq 0.11 0.72 0.2 0.9
gamma 0.05 0.99 0.0 1.0
log likelihood value: -18.46
cross-sectional data
total number of observations = 60
mean efficiency: 1.0617
>
> ## L(MLE) < L(OLS)
> a4 <- sfa( log( output ) ~ log( capital ) + log( labour ),
+ data = front41Data, start = c( 1, 0, 0, 1, 0.999995 ) )
Warning messages:
1: In sfa(log(output) ~ log(capital) + log(labour), data = front41Data, :
the likelihood value of the ML estimation is less than that obtained using OLS; this indicates that the likelihood maximization did not converge to the global maximum or that there is no inefficiency (you could try again using different starting values)
2: In sfa(log(output) ~ log(capital) + log(labour), data = front41Data, :
the parameter 'gamma' is close to the boundary of the parameter space [0,1]: this can cause convergence problems and can negatively affect the validity and reliability of statistical tests and might be caused by model misspecification
> print( summary( a4 ), digits = 1 )
Error Components Frontier (see Battese & Coelli 1992)
Inefficiency decreases the endogenous variable (as in a production function)
The dependent variable is logged
Iterative ML estimation terminated after 10 iterations:
cannot find a parameter vector that results in a log-likelihood value
larger than the log-likelihood value obtained in the previous step
final maximum likelihood estimates
Estimate Std. Error z value Pr(>|z|)
(Intercept) -6e+00 4e+00 -1.5 0.1
log(capital) 7e-01 9e-01 0.7 0.5
log(labour) 1e+00 9e-01 1.6 0.1
sigmaSq 3e+01 3e+00 12.4 <2e-16 ***
gamma 1e-08 3e-06 0.0 1.0
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
log likelihood value: -162.91
cross-sectional data
total number of observations = 60
mean efficiency: 0.99954
>
> ## too many starting values
> try( sfa( log( output) ~ log( capital ) + log( labour ), data = front41Data,
+ truncNorm = TRUE, startVal = c( 0.5, 0.3, 0.5, 0.5, 0.9, -1, 0.3 ) ) )
Error in sfa(log(output) ~ log(capital) + log(labour), data = front41Data, :
wrong number of starting values (you provided 7 starting values but the model has 6 parameters)
>
> ## too few starting values
> try( sfa( log( output) ~ log( capital ) + log( labour ), data = front41Data,
+ truncNorm = TRUE, startVal = c( 0.5, 0.3, 0.5, 0.5, 0.9 ) ) )
Error in sfa(log(output) ~ log(capital) + log(labour), data = front41Data, :
wrong number of starting values (you provided 5 starting values but the model has 6 parameters)
>
> ## one explanatory variable specifiec twice (works)
> sfa( log( output) ~ log( capital ) + log( labour ) + log( capital ),
+ data = front41Data )
Call:
sfa(formula = log(output) ~ log(capital) + log(labour) + log(capital), data = front41Data)
Maximum likelihood estimates
(Intercept) log(capital) log(labour) sigmaSq gamma
0.562 0.281 0.536 0.217 0.797
>
> ## perfect multicollinearity -> NAs in OLS coefficiencts
> front41Data$capital10 <- 10 * front41Data$capital
> try( sfa( log( output) ~ log( capital ) + log( labour ) + log( capital10 ),
+ data = front41Data ) )
Error in sfa(log(output) ~ log(capital) + log(labour) + log(capital10), :
at least one coefficient estimated by OLS is NA: log(capital10). This may have been caused by (nearly) perfect multicollinearity
>
> ## perfect multicollinearity -> 2 NAs in OLS coefficiencts
> front41Data$capitalLabour <- front41Data$capital * front41Data$labour
> try( sfa( log( output) ~ log( capital ) + log( labour ) + log( capital10 ) +
+ log( capitalLabour ), data = front41Data ) )
Error in sfa(log(output) ~ log(capital) + log(labour) + log(capital10) + :
at least one coefficient estimated by OLS is NA: log(capital10), log(capitalLabour). This may have been caused by (nearly) perfect multicollinearity
>
>
> ## load data abour rice production in the Phillipines
> data( "riceProdPhil")
>
> ## nobs > nn * nt
> rd <- riceProdPhil
> rd <- rbind( rd, rd[ 11, ] )
> rd <- pdata.frame( rd, c( "FMERCODE", "YEARDUM" ), row.names = FALSE )
Warning message:
In pdata.frame(rd, c("FMERCODE", "YEARDUM"), row.names = FALSE) :*** buffer overflow detected ***: terminated
Aborted
Flavor: r-devel-linux-x86_64-debian-gcc