performance: Transfer function model performance statistics

In analogue: Analogue and Weighted Averaging Methods for Palaeoecology

Description

A simple extractor function to access the model performance statistics of transfer function models.

Usage

performance(object, ...)

Arguments

object	A transfer function object.
...	Arguments passed to other methods. Currently ignored.

Details

performance is a generic function for use with a number of fitted models objects in analogue. The available methods are:

• wa Weighted Averaging Models.

• predict.wa Predictions from a Weighted Average Model.

• pcr Principal Component Regression models.

• bootstrap.wa Bootstrapped Weighted Averaging Models.

• crossval Cross-validated models fitted via crossval.

Value

A named vector containing the extracted model performance statistics.

Author(s)

Gavin L. Simpson

See Also

wa, predict.wa, bootstrap.wa.

Examples

data(ImbrieKipp)
data(SumSST)

## fit the WA model
mod <- wa(SumSST ~., data = ImbrieKipp)
mod

## the model performance statistics
performance(mod)

Example output

Loading required package: vegan
Loading required package: permute
Loading required package: lattice
This is vegan 2.5-7
analogue version 0.17-5

	Weighted Averaging Transfer Function

Call:
wa(formula = SumSST ~ ., data = ImbrieKipp) 

Deshrinking  : Inverse 
Tolerance DW : No 
No. samples  : 61 
No. species  : 27 

Performance:
     RMSE  R-squared  Avg. Bias  Max. Bias  
   2.0188     0.9173     0.0000    -3.8155  

    RMSE       R2 Avg.Bias Max.Bias 
   2.019    0.917    0.000   -3.815 

analogue documentation built on June 21, 2021, 1:08 a.m.