Statistics

mne.stats:

Functions for statistical analysis.

Parametric statistics (see scipy.stats and statsmodels for more options):

ttest_1samp_no_p(X[, sigma, method])	Perform one-sample t-test.
ttest_ind_no_p(a, b[, equal_var, sigma])	Independent samples t-test without p calculation.
f_oneway(*args)	Perform a 1-way ANOVA.
f_mway_rm(data, factor_levels[, effects, ...])	Compute M-way repeated measures ANOVA for fully balanced designs.
f_threshold_mway_rm(n_subjects, factor_levels)	Compute F-value thresholds for a two-way ANOVA.
linear_regression(inst, design_matrix[, names])	Fit Ordinary Least Squares (OLS) regression.
linear_regression_raw(raw, events[, ...])	Estimate regression-based evoked potentials/fields by linear modeling.

Mass-univariate multiple comparison correction:

bonferroni_correction(pval[, alpha])	P-value correction with Bonferroni method.
fdr_correction(pvals[, alpha, method])	P-value correction with False Discovery Rate (FDR).

Non-parametric (clustering) resampling methods:

combine_adjacency(*structure)	Create a sparse binary adjacency/neighbors matrix.
permutation_cluster_test(X[, threshold, ...])	Cluster-level statistical permutation test.
permutation_cluster_1samp_test(X[, ...])	Non-parametric cluster-level paired t-test.
permutation_t_test(X[, n_permutations, ...])	One sample/paired sample permutation test based on a t-statistic.
spatio_temporal_cluster_test(X[, threshold, ...])	Non-parametric cluster-level test for spatio-temporal data.
spatio_temporal_cluster_1samp_test(X[, ...])	Non-parametric cluster-level paired t-test for spatio-temporal data.
summarize_clusters_stc(clu[, p_thresh, ...])	Assemble summary SourceEstimate from spatiotemporal cluster results.
bootstrap_confidence_interval(arr[, ci, ...])	Get confidence intervals from non-parametric bootstrap.

ERP-related statistics:

erp.compute_sme(epochs[, start, stop])

Compute standardized measurement error (SME).

Compute adjacency matrices for cluster-level statistics:

channels.find_ch_adjacency(info, ch_type)	Find the adjacency matrix for the given channels.
channels.read_ch_adjacency(fname[, picks])	Read a channel adjacency ("neighbors") file that ships with MNE.
spatial_dist_adjacency(src, dist[, verbose])	Compute adjacency from distances in a source space.
spatial_src_adjacency(src[, dist, verbose])	Compute adjacency for a source space activation.
spatial_tris_adjacency(tris[, ...])	Compute adjacency from triangles.
spatial_inter_hemi_adjacency(src, dist[, ...])	Get vertices on each hemisphere that are close to the other hemisphere.
spatio_temporal_src_adjacency(src, n_times)	Compute adjacency for a source space activation over time.
spatio_temporal_tris_adjacency(tris, n_times)	Compute adjacency from triangles and time instants.
spatio_temporal_dist_adjacency(src, n_times, ...)	Compute adjacency from distances in a source space and time instants.