mne.stats.ttest_1samp_no_p

mne.stats.ttest_1samp_no_p(X, sigma=0, method=’relative’)[source]

Perform t-test with variance adjustment and no p-value calculation.

Parameters:

X : array

Array to return t-values for.

sigma : float

The variance estate will be given by “var + sigma * max(var)” or “var + sigma”, depending on “method”. By default this is 0 (no adjustment). See Notes for details.

method : str

If ‘relative’, the minimum variance estimate will be sigma * max(var), if ‘absolute’ the minimum variance estimate will be sigma.
Returns:

t : array

t-values, potentially adjusted using the hat method.

Notes

One can use the conversion:

threshold = -scipy.stats.distributions.t.ppf(p_thresh, n_samples - 1)

to convert a desired p-value threshold to t-value threshold. Don’t forget that for two-tailed tests, p_thresh in the above should be divided by 2.

To use the “hat” adjustment method, a value of sigma=1e-3 may be a reasonable choice. See Ridgway et al. 2012 “The problem of low variance voxels in statistical parametric mapping; a new hat avoids a ‘haircut’”, NeuroImage. 2012 Feb 1;59(3):2131-41.

Examples using mne.stats.ttest_1samp_no_p