extract.autopls: Extract information from a fitted autopls model

In autopls: Partial Least Squares Regression with Backward Selection of Predictors

Description

Functions to extract information from autopls objects: crossvalidation, fitted values, regression coefficients, residuals, scores, loadings, latent vectors used, underlying run.

Usage

  predicted (object)
  get.lv (object)
  get.iter (object)
  slim (object)
  ## S3 method for class 'autopls'
scores(object, ...)
  ## S3 method for class 'autopls'
loadings(object, ...)
  ## S3 method for class 'autopls'
fitted(object, ...)
  ## S3 method for class 'autopls'
coef(object, intercept = FALSE, ...)
  ## S3 method for class 'slim'
coef(object, intercept = FALSE, ...)
  ## S3 method for class 'autopls'
residuals(object, ...)

Arguments

object	object of class autopls
intercept	logical. Should intercept be given?
...	logical. Arguments to be passed to methods

Details

Provides convenience wrappers for extract functions in package pls. More details are given here: coef.mvr. Other functions extract information specific for autopls objects: get.lv, get.iter or condense the model information to a memory saving object of class slim that can be used for predictions with predict.slim. This makes sense if large pedictor data sets result in huge autopls model objects that are difficult to handle.

Value

see coef.mvr. get.iter returns the run in the autopls backwards selection procedure that has been used for the current model. get.lv returns the number of latent vectors used for the present model. predicted returns the predictions in model validation while fitted returns the predictions in model calibration. slim returns an object of class slim.

Note

If you want to make full use of the extract functions in the pls package assign class mvr to the model object.

Reducing a model to an object of class slim means loosing plotting options.

Author(s)

Sebastian Schmidtlein, links to code from package pls by Ron Wehrens and Bjørn-Helge Mevik.

See Also

autopls, metaval, set.iter, set.lv, predict.slim

Examples

  ## load predictor and response data to the current environment
  data (murnau.X)
  data (murnau.Y)
  
  ## call autopls with the standard options
  model <- autopls (murnau.Y ~ murnau.X)
  
  ## get fitted values
  fitted(model)

Example output

Loading required package: pls

Attaching package: 'pls'

The following object is masked from 'package:stats':

    loadings

autopls 1.3
1   Pred: 26  LV: 3   R2v: 0.74   RMSEv: 4.727  
2   Pred: 23  LV: 3   R2v: 0.742  RMSEv: 4.705  Criterion: A1
3   Pred: 20  LV: 3   R2v: 0.749  RMSEv: 4.645  Criterion: A4
4   Pred: 18  LV: 3   R2v: 0.752  RMSEv: 4.611  Criterion: A4
5   Pred: 16  LV: 3   R2v: 0.752  RMSEv: 4.61   Criterion: A4
6   Pred: 13  LV: 3   R2v: 0.76   RMSEv: 4.537  Criterion: A1
7   Pred: 11  LV: 3   R2v: 0.768  RMSEv: 4.466  Criterion: A4
8   Pred: 9   LV: 3   R2v: 0.775  RMSEv: 4.397  Criterion: A4

Predictors: 9   Observations: 40   Latent vectors: 3   Run: 8 
RMSE(CAL): 4.09   RMSE(LOO): 4.4   
R2(CAL): 0.805    R2(LOO): 0.775   
      s01       s02       s03       s04       s05       s07       s09       s10 
 4.020513  6.075465  8.468439 14.724063  3.432336 13.895287 10.053383 20.500551 
      s11       s12       s14       s15       s16       s17       s19       s20 
10.436921 11.572937 13.177720 20.045069 22.393698 27.597733 12.668839 17.797782 
      s21       s22       s23       s24       s25       s26       s27       s28 
26.327236 28.398754  4.594793 10.735778 11.310073 25.195901 26.718381 30.000960 
      s29       s30       s31       s32       s33       s34       s35       s36 
14.340586 15.506982 26.734095 26.890262 29.420807 20.688843 23.219387 14.885150 
      s37       s38       s39       s40       s41       s42       s43       s44 
26.853603 27.738141 30.604164 24.174041 20.830374 28.341443 27.849402 30.811998 

autopls documentation built on May 2, 2019, 9:39 a.m.