Empirical Likelihood for General Block Designs

Performs hypothesis testing for general block designs with empirical likelihood. The core computational routines are implemented using the 'Eigen' 'C++' library and 'RcppEigen' interface, with 'OpenMP' for parallel computation. Details of the methods are given in Kim, MacEachern, and Peruggia (2023) . This work was supported by the U.S. National Science Foundation under Grants No. SES-1921523 and DMS-2015552.

elgbd

Overview

elgbd performs hypothesis testing for general block designs with empirical likelihood. The core computational routines are implemented with the ‘Eigen’ ‘C++’ library and ‘RcppEigen’ interface, with ‘OpenMP’ for parallel computation. Details of the testing procedures are given in Kim, MacEachern, and Peruggia (2023). This work was supported by the U.S. National Science Foundation under Grants No. SES-1921523 and DMS-2015552.

Installation

You can install the latest stable release of elgbd from CRAN.

install.packages("elgbd")

Development version

You can install the development version of elgbd from GitHub.

# install.packages("pak")
pak::pak("markean/elgbd")

Usage

library(elgbd)
set.seed(58234)
# Analysis of variance
el_aov(formula = Sepal.Length ~ Species, data = iris)
#> Call:
#> el_aov(formula = Sepal.Length ~ Species, data = iris)
#> 
#> Minimizer:
#> 5.9595 5.9595 5.9595
#> 
#> Statistic:
#> 94.9054

# All pairwise comparisons
data("clothianidin")
el_pairwise(clo ~ trt | blk, data = clothianidin, B = 1000)
#> 
#>  Empirical Likelihood Multiple Tests
#> 
#> All pairwise comparisons
#> 
#>                   Estimate  Chisq  Lwr.ci  Upr.ci  p.adj    
#> Naked - Fungicide  -1.0525 14.098 -2.1150 -0.2794  0.002 ** 
#> Naked - Low        -1.6794 15.935 -2.6705 -0.6435  0.001 ***
#> Naked - High       -3.1726 42.803 -4.5731 -1.6940 <0.001 ***
#> Fungicide - Low    -0.6269  1.834 -1.6303  0.6448  0.523    
#> Fungicide - High   -2.1201 14.085 -3.2313 -0.7247  0.002 ** 
#> Low - High         -1.4932 11.715 -2.7228 -0.3954  0.003 ** 
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> k: 1, level: 0.05, method: AMC, cutoff: 6.6737

# Comparisons with control
el_pairwise(clo ~ trt | blk,
  data = clothianidin, control = "Naked", method = "NB", B = 1000
)
#> 
#>  Empirical Likelihood Multiple Tests
#> 
#> Comparisons with control
#> 
#>                   Estimate Chisq Lwr.ci Upr.ci p.adj   
#> Fungicide - Naked   1.0525 14.10 0.3506 1.9900 0.005 **
#> Low - Naked         1.6794 15.94 0.7586 2.5646 0.004 **
#> High - Naked        3.1726 42.80 1.8489 4.4282 0.003 **
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> k: 1, level: 0.05, method: NB, cutoff: 6.3291

# Test for equal means
el_test(clo ~ trt | blk, clothianidin,
  lhs = matrix(c(
    1, -1, 0, 0,
    0, 1, -1, 0,
    0, 0, 1, -1
  ), byrow = TRUE, nrow = 3L)
)
#> 
#>  Empirical Likelihood Test
#> 
#> General block designs
#> 
#> Maximum EL estimates:
#> [1] -4.479086 -3.426618 -2.799700 -1.306530
#> 
#> Statistic: 28.055