mean_comparisons.predict_the_past_model_bh_GxE: Get mean comparisons from 'predict_the_past_model_bh_GxE'...
In priviere/PPBstats: An R package to perform analysis found within PPB programmes

Description Usage Arguments Details Value Author(s) See Also

View source: R/mean_comparisons.predict_the_past_model_bh_GxE.R

mean_comparisons performs mean comparisons from object coming from predict_the_past_model_bh_GxE

## S3 method for class 'predict_the_past_model_bh_GxE'
mean_comparisons(x, alpha = 0.05,
  type = 1, get.at.least.X.groups = 2, precision = 5e-04,
  threshold = 1, p.adj = "soft.bonf", ...)

`x`	outputs from `predict_the_past_model_bh_GxE`
`alpha`	level of type one error. 0.05 (5%) by default
`type`	type of comparisons type = 1 for comparison two by two type = 2 for comparison to a specific threshold
`get.at.least.X.groups`	For type = 1. If there are only one group with alpha, the minimum number of groups wanted with a higher type one error (i.e. lower confidence). If NULL, nothing is done.
`precision`	For type = 1. The precision of the alpha with the correspondong groups from get.at.least.X.groups. The smaller the better, but the smaller the more time consuming due to computing matters
`threshold`	For type = 2. The threshold to which a parameter is different
`p.adj`	For all except type = 2. NULL for no adjustement of the type one error. p.adj can be "soft.bonf". p.adj = "soft.bonf" for a soft bonferonni correction to take into account multiple comparisons (alpha / nb of parameters).. The comparisons is based on the probability of having a common distribution for each pair of parameter. When there is only one group with the value of alpha, the function (via `get.at.least.X.groups argument`) returns at least X groups with a new value of alpha.
`...`	further arguments passed to or from other methods#'

S3 method. For more details, see in the book : https://priviere.github.io/PPBstats_book/intro-agro.html#section-bayes

A list of one element : data_mean_comparisons, composed of a list of one element for the given environment choose, being a list of two elements:

mean.comparisons: a dataframe with the following columns : parameter, median, groups, number of groups, alpha (type one error), alpha.correction (correction used), entry, environment, location and year.
Mpvalue : a square matrix with pvalue computed for each pair of parameter.

parameter is set to "mu" by default

Pierre Riviere