plspm (version 0.2-2)

plsreg2: PLS-R2: Partial Least Squares Regression 2

Description

Calculates partial least squares regression for the multivariate case (i.e. more than one response variable)

Usage

plsreg2(X, Y, nc = 2)

Arguments

X

A numeric matrix or data frame containing the predictor variables.

Y

A numeric matrix or data frame containing the predictand variables.

nc

The number of extracted PLS components (2 by default)

Value

• An object of class "plsreg2", basically a list with the following elements:
• x.scores
  components of the predictor variables.
• x.loads
  loadings of the predictor variables.
• y.scores
  components of the predictand variables.
• y.loads
  loadings of the predictand variables.
• raw.wgs
  weights to calculate the PLS scores with the deflated matrices of predictor variables.
• mod.wgs
  modified weights to calculate the PLS scores with the matrix of predictor variables.
• cor.tx
  modified weights to calculate the PLS scores with the matrix of predictor variables.
• cor.ty
  modified weights to calculate the PLS scores with the matrix of predictor variables.
• std.coef
  Vector of standardized regression coefficients.
• coeffs
  Vector of regression coefficients (used with the original data scale).
• y.pred
  Vector of predicted values.
• resid
  Vector of residuals.
• expvar
  table with R-squared coefficients.
• Q2
  table of Q2 indexes (i.e. leave-one-out cross validation).
• Q2cum
  table of cummulated Q2 indexes.
• VIP
  Variable Importance for Projection.

Details

The minimum number of PLS components nc is 2. The data is scaled to standardized values (mean=0, variance=1). No missing data are allowed. Argument Y must contain more than one variable. If Y is a vector, you may use the function plsreg1.

References

Geladi, P., and Kowlaski, B. (1986) Partial Least Squares Regression: A Tutorial. Analytica Chimica Acta, 185, pp. 1-17. Hoskuldsson, A. (1988) PLS Regression Methods. Journal of Chemometrics, 2, pp. 211-228. Tenenhaus, M. (1998) La Regression PLS. Theorie et Pratique. Editions TECHNIP, Paris. Valencia, J.L., Diaz-Llanos, F.J. (2004) Metodos de Prediccion en Situaciones Limite. Editorial La Muralla, S.A. Madrid.

See Also

print.plsreg2, plot.plsreg2, plsreg1.

Examples

## example of PLSR2 with the vehicles dataset
  data(vehicles)
  pls2 <- plsreg2(vehicles[,1:12], vehicles[,13:16])
  pls2
  plot(pls2)