The R package melt provides a unified framework for data analysis with empirical likelihood methods. A collection of functions are available for basic regression analysis and hypothesis testing. Much of its functionality and syntax mimics the corresponding base R functions. The core computational routines are implemented with the ‘Eigen’ C++ library and ‘RcppEigen’ interface, with OpenMP for parallel computation. Additional functions are available for multiple testing for the analysis of experimental designs. Details of the testing procedures are given in Kim et al. (2021).
You can install the latest stable release from CRAN.
install.packages("melt")You can install the latest development version from Github.
# install.packages("devtools")
devtools::install_github("markean/melt")library(melt)
# one sample test for mean
fit1 <- el_mean(par = 0, x = rnorm(n = 100))
confint(fit1)
#> lower upper
#> 1 -0.3927377 0.01727951
# linear regression
fit2 <- el_lm(formula = mpg ~ wt, data = mtcars)
summary(fit2)
#>
#> Call:
#> el_lm(formula = mpg ~ wt, data = mtcars)
#>
#> Coefficients:
#> Estimate Chisq Pr(>Chisq)
#> (Intercept) 37.285 443.3 <2e-16 ***
#> wt -5.344 439.1 <2e-16 ***
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Chisq: 439.1, df: 1, Pr(>Chisq): < 2.2e-16
#>
#> Constrained EL: not converged
confint(fit2)
#> lower upper
#> (Intercept) 33.175865 41.865426
#> wt -6.746902 -4.149511
# analysis of variance
data("clothianidin")
fit3 <- el_lm(clo ~ -1 + trt, clothianidin)
summary(fit3)
#>
#> Call:
#> el_lm(formula = clo ~ -1 + trt, data = clothianidin)
#>
#> Coefficients:
#> Estimate Chisq Pr(>Chisq)
#> trtNaked -4.479 411.072 < 2e-16 ***
#> trtFungicide -3.427 59.486 1.23e-14 ***
#> trtLow -2.800 62.955 2.11e-15 ***
#> trtHigh -1.307 4.653 0.031 *
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Chisq: 894.4, df: 4, Pr(>Chisq): < 2.2e-16
#>
#> Constrained EL: not converged
confint(fit3)
#> lower upper
#> trtNaked -5.002118 -3.9198229
#> trtFungicide -4.109816 -2.6069870
#> trtLow -3.681837 -1.9031795
#> trtHigh -2.499165 -0.1157222
# test of no treatment effect
lhs <- matrix(c(1, -1, 0, 0,
0, 1, -1, 0,
0, 0, 1, -1), byrow = TRUE, nrow = 3)
lht(fit3, lhs = lhs)
#>
#> Empirical Likelihood: lm
#>
#>
#> Chisq: 26.6, df: 3, Pr(>Chisq): 7.148e-06
#>
#> EL evaluation: converged
# logistic regression
fit4 <- el_glm(vs ~ mpg, data = mtcars, family = binomial)
summary(fit4)
#>
#> Call:
#> el_glm(formula = vs ~ mpg, family = binomial, data = mtcars)
#>
#> Coefficients:
#> Estimate Chisq Pr(>Chisq)
#> (Intercept) -8.8331 24.79 6.39e-07 ***
#> mpg 0.4304 24.42 7.75e-07 ***
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Chisq: 24.42, df: 1, Pr(>Chisq): 7.751e-07
#>
#> Constrained EL: converged