neuralnet: confidence.interval – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

confidence.interval

Calculates confidence intervals of the weights

Description

confidence.interval, a method for objects of class nn, typically produced by neuralnet. Calculates confidence intervals of the weights (White, 1989) and the network information criteria NIC (Murata et al. 1994). All confidence intervals are calculated under the assumption of a local identification of the given neural network. If this assumption is violated, the results will not be reasonable. Please make also sure that the chosen error function equals the negative log-likelihood function, otherwise the results are not meaningfull, too.

Usage

confidence.interval(x, alpha = 0.05)

Arguments

`x`	neural network
`alpha`	numerical. Sets the confidence level to (1-alpha).

Value

confidence.interval returns a list containing the following components:

`lower.ci`	a list containing the lower confidence bounds of all weights of the neural network differentiated by the repetitions.
`upper.ci`	a list containing the upper confidence bounds of all weights of the neural network differentiated by the repetitions.
`nic`	a vector containg the information criteria NIC for every repetition.

Author(s)

Stefan Fritsch, Frauke Guenther guenther@leibniz-bips.de

References

White (1989) Learning in artificial neural networks. A statistical perspective. Neural Computation (1), pages 425-464

Murata et al. (1994) Network information criterion - determining the number of hidden units for an artificial neural network model. IEEE Transactions on Neural Networks 5 (6), pages 865-871

Examples

data(infert, package="datasets")
    print(net.infert <- neuralnet(case~parity+induced+spontaneous,  
                        infert, err.fct="ce", linear.output=FALSE))
    confidence.interval(net.infert)

neuralnet

Training of Neural Networks

v1.44.2

GPL (>= 2)

Authors

Stefan Fritsch [aut], Frauke Guenther [aut], Marvin N. Wright [aut, cre], Marc Suling [ctb], Sebastian M. Mueller [ctb]

Initial release

2019-02-07