Skip to contents

Configure PLS Analysis

Source: R/builder_config.R
configure.Rd

Sets analysis parameters for the PLS specification.

Usage

configure(
  spec,
  method = NULL,
  nperm = NULL,
  nboot = NULL,
  nsplit = NULL,
  clim = NULL,
  meancentering = NULL,
  cormode = NULL,
  boot_type = NULL,
  is_struct = NULL
)

Arguments

spec

A pls_spec object

method

PLS method (character or integer):

  • "task" or 1: Mean-Centering Task PLS

  • "task_nonrotated" or 2: Non-Rotated Task PLS

  • "behavior" or 3: Regular Behavior PLS

  • "multiblock" or 4: Regular Multiblock PLS

  • "behavior_nonrotated" or 5: Non-Rotated Behavior PLS

  • "multiblock_nonrotated" or 6: Non-Rotated Multiblock PLS

  • "ws_seed": Alias for task PLS on trial-level within-subject seed correlation maps from add_trial_data()

  • "ws_seed_nonrotated": Alias for non-rotated task PLS on trial-level within-subject seed correlation maps

nperm

Number of permutations (0 = no permutation test)

nboot

Number of bootstrap samples (0 = no bootstrap)

nsplit

Number of split-half iterations (0 = no split-half)

clim

Confidence level for bootstrap (0-100, default 95)

meancentering

Mean-centering type (0-3):

  • 0: Within-group centering (default)

  • 1: Grand condition mean centering

  • 2: Grand mean centering

  • 3: Remove all main effects

cormode

Correlation mode for behavior PLS:

  • "pearson" or 0: Pearson correlation (default)

  • "covariance" or 2: Covariance

  • "cosine" or 4: Cosine angle

  • "dot" or 6: Dot product

boot_type

Bootstrap type: "strat" (default) or "nonstrat"

is_struct

Logical, structure PLS (don't permute conditions)

Value

Updated pls_spec object

Examples

set.seed(42)
data1 <- matrix(rnorm(60 * 50), 60, 50)
data2 <- matrix(rnorm(54 * 50), 54, 50)

spec <- pls_spec() |>
  add_subjects(list(data1, data2), groups = c(20, 18)) |>
  add_conditions(3) |>
  configure(method = "task", nperm = 100, nboot = 50)