EffectsRelBaseline-package {EffectsRelBaseline}R Documentation

Test changes of a grouped response relative to baseline.


Functions to test for changes of a response to a stimulus grouping relative to a background or baseline response.


Package: EffectsRelBaseline
Version: 0.4
Date: 2013-06-12
License: GPL version 3 or newer
Built: R 2.15.2; ; 2013-06-12 22:41:44 UTC; unix


ef          Example data of human single neuron responses to presentation of emotional faces.
                        F-ratio statistic for grouped responses.
                        Log-likelihood ratio statistic for grouped responses.
                        Log-likelihood ratio of grouped responses being drawn from 
                        a background distribution with known mean and variance.
                        Weighted sum of squared deviations test statistic, as used
                        in the changes from background test (CBT).

This package provides functions for testing whether grouped responses have significant deviations from baseline. It allows testing using the non-parametric changes from background test (CBT), which doesn't require multiple testing relative to baseline followed by a second test for an effect of group on the responses. Such double testing can lead to erroneous conclusions, as discussed in Steinmetz & Thorp, 2013.

The primary function is sumSqCat which computes the appropriate test statistic for use in the CBT. A variety of other test functions are provided which can be used for comparison of performance.


Peter N. Steinmetz <PeterNSteinmetz@steinmetz.org> Christopher Thorp <thorp@spacia.org>

Maintainer: Peter N. Steinmetz <PeterNSteinmetz@steinmetz.org>


Steinmetz, P.N. & Thorp, C.K. (2013) Testing for effects of different stimuli on neuronal firing relative to background activity. Journal of Neural Engineering, Sept. 2013.

See Also

PsumtSim-package boot

[Package EffectsRelBaseline version 0.5 Index]