bootstrap: bootstrap function
In nsantantonio/Bilinear: Fits bilinear models for multi environment trial data

Description Usage Arguments Details Value

This function perfoms a bootstrap test on one PC at a time inside bilinear(), for ease of parallelization. Not meant to be used by user.

1 2	bootstrap(K, D, Dtilde, Edecomp, M, B, model, bootMethod = "full", Theta_k = NULL, verbose = FALSE, ...)

`K`	integer, kth eigenvalue
`D`	integer, Maximum right dimensions
`Dtilde`	integer. Maximum right dimensions
`Edecomp`	svd() object from singular value decomposition of the table of residuals (thetaPlusR)
`M`	integer. maximum number of PCs.
`B`	integer. Number of bootstraps to be performed.
`model`	bilinear model to be fit. Arguments can be "AMMI", "GGE", "SREG", "EGE", "GREG". "GGE" and "SREG" are equivalent, as are "EGE" and "GREG".
`bootMethod`	character. Method for bootstrap sampling. Can be "full" or "simple", default is "full".
`Theta_k`	Theta the reduced dimension table can be provided directly. useful for ??
`verbose`	logical. Should details be printed?
`...`	Additional arguments.

Performs bootstrap test of significance by returning the proportion of time the bootstrap test statistic exceeds the statistic calculated from the Kth principal component. See Forkman and Piepho (2014) Biometrics, 70(3) for specifics.

p-values for 1 to K principal components

nsantantonio/Bilinear documentation built on Aug. 18, 2020, 2:31 p.m.