mbclusterwise (version 1.0)

mbregular: Regularized multiblock regression

Description

Function to perform the regularized multiblock regression which gives results comprised the ones from multiblock Redundancy Analysis (gamma=0) and multiblock PLS (gamma=1). This method is applied to several explanatory blocks $(X_1, \dots, X_K)$ defined as an object of class ktab (from ade4), to explain a dependent dataset $Y$ defined as an object of class dudi (from ade4).

Usage

mbregular(dudiY, ktabX, scale = FALSE, option = c("none", "uniform"), H, gamma)

Arguments

dudiY

an object of class dudi (from ade4) containing the dependent variable(s)

ktabX

an object of class ktab (from ade4) containing the blocks of explanatory variables

scale

a logical value indicating whether the explanatory variables should be standardized

option

an option for the block weighting (by default, the first option is chosen): none the block weight is equal to the block inertia uniform the block weight is equal to $1/K$ for $(X_1, \dots, X_K)$ and to 1 for $X$ and $Y$

H

an integer giving the number of dimensions

gamma

a numeric value of the regularization parameter comprised between 0 and 1. The value (gamma=0) leads to multiblock Redundancy Analysis and (gamma=1) to multiblock PLS

Value

References

Bougeard, S., Qannari, E.M., Lupo, C. and Hanafi, M. (2011). From multiblock partial least squares to multiblock redundancy analysis. A continuum approach. Informatica, 22(1), 11-26

See Also

cw.multiblock, cw.tenfold, cw.predict, mbpcaiv, mbpls

Examples

  data(simdata.red) 
  Data.X <- simdata.red[c(1:15, 21:35), 1:10]
  Data.Y <- simdata.red[c(1:15, 21:35), 11:13]
  library(ade4)
  dudiy  <- dudi.pca(df = Data.Y, center = FALSE, scale = FALSE, scannf = FALSE)
  ktabx  <- ktab.data.frame(df = data.frame(Data.X), blocks = c(5,5), 
   tabnames = paste("Tab", c(1:2), sep = "."))
  res    <- mbregular(dudiy, ktabx, scale = FALSE, option = "none", H = 2, gamma = 0.8)