Pairwise Tukey Comparisons — pairwise

pairwise_comparison computes pairwise Tukey comparisons for any combination of variables in a given QGAM using the emmeans package. Pairwise Tukey comparisons are most sensible to check differences between levels of a categorical predictor.

pairwise_comparison(
  qgam,
  quantile = NULL,
  factor, by = NULL,
  ...)

Arguments

qgam: A qgam object created with qgam::qgam or extracted from a qgam::mqgam object, or a collection of qgams created with qgam::mqgam.
quantile: If qgam is a collection of qgam models, specify the quantile you are interested in. Not meaningful for single qgam objects.
factor: The categorical variable you wish to compute Tukey comparisons for. Must be one of the predictor or smooth terms given in the qgam formula.
...: Any of the other arguments emmeans::emmeans can take.

References

Russell V. Lenth (2021). emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version 1.5.4. https://CRAN.R-project.org/package=emmeans

Author

D. Schmitz

Examples


pairwise_comparison(qgam = mtqgam_mqgam, quantile = 0.5, "factor_1")
#>  contrast estimate   SE   df t.ratio p.value
#>  J - K        0.12 1.54 4984  0.078  0.9998 
#>  J - L       -3.52 6.81 4984 -0.517  0.9550 
#>  J - M       -1.01 1.62 4984 -0.620  0.9257 
#>  K - L       -3.64 6.67 4984 -0.545  0.9478 
#>  K - M       -1.13 0.92 4984 -1.225  0.6110 
#>  L - M        2.51 6.69 4984  0.376  0.9819 
#> 
#> Results are averaged over the levels of: factor_2, factor_3 
#> P value adjustment: tukey method for comparing a family of 4 estimates