Table Of Contents

Next topic

Automatic plotting of activation maps

This Page

Empirical null

The fff2.utils.emp_null module contains a class that fits a gaussian model to the central part of an histogram, following Schwartzman et al, 2009. This is typically necessary to estimate a fdr when one is not certain that the data behaves as a standard normal under H_0.

The ENN class learns its null distribution on the data provided at initialisation. Two different methods can be used to set a threshold from the null distribution: the ENN.threshold() method returns the threshold for a given false discovery rate, and thus accounts for multiple comparisons with the given dataset; the ENN.uncorrected_threshold() returns the threshold for a given uncorrected p-value, and as such does not account for multiple comparisons.

Example

If we use the empirical normal null estimator on a two gaussian mixture distribution, with a central gaussian, and a wide one, it uses the central distribution as a null hypothesis, and returns the threshold followingr which the data can be claimed to belong to the wide gaussian:

import numpy as np
x = np.c_[np.random.normal(size=1e4),
          np.random.normal(scale=4, size=1e4)]

from nipy.neurospin.utils.emp_null import ENN
enn = ENN(x)
enn.threshold(verbose=True)

[source code, hires.png, pdf]

../_images/enn_demo.png

The threshold evaluated with the ENN.threshold() method is around 2.8 (using the default p-value of 0.05). The ENN.uncorrected_threshold() return, for the same p-value, a threshold of 1.9. It is necessary to use a higher p-value with uncorrected comparisons.

Class documentation


Reference: Schwartzmann et al., NeuroImage 44 (2009) 71–82