API

pymare.core: Core objects

Core classes and functions.

core.Dataset([y, v, X, n, data, X_names, ...])	Container for input data and arguments to estimators.
core.meta_regression([y, v, X, n, data, ...])	Fit the standard meta-regression/meta-analysis model to provided data.

pymare.estimators: Meta-analytic algorithms

Estimators for meta-analyses and meta-regressions.

estimators.WeightedLeastSquares([tau2, ...])	Weighted least-squares meta-regression.
estimators.DerSimonianLaird([weight_scheme, ...])	DerSimonian-Laird meta-regression estimator.
estimators.VarianceBasedLikelihoodEstimator([...])	Likelihood-based estimator for estimates with known variances.
estimators.SampleSizeBasedLikelihoodEstimator([...])	Likelihood-based estimator for data with known sample sizes but unknown sampling variances.
estimators.StanMetaRegression([tau_prior_scale])	Bayesian meta-regression estimator using Stan.
estimators.Hedges([weight_scheme, rho, ...])	Hedges meta-regression estimator.
estimators.StoufferCombinationTest([mode, ...])	Stouffer's Z-score meta-analysis method.
estimators.FisherCombinationTest([mode])	Fisher's method for combining p-values.
estimators.estimators.BaseEstimator()	A base class for Estimators.

pymare.results: Meta-analytic results

Tools for representing and manipulating meta-regression results.

results.MetaRegressionResults(estimator, ...)	Container for results generated by PyMARE meta-regression estimators.
results.CombinationTestResults(estimator, ...)	Container for results generated by p-value combination methods.
results.PermutationTestResults(results, ...)	Lightweight container to hold and display permutation test results.
results.BayesianMetaRegressionResults(data, ...)	Container for MCMC sampling-based PyMARE meta-regression estimators.

pymare.effectsize: Effect size computation/conversion

Tools for converting between effect-size measures.

effectsize.OneSampleEffectSizeConverter([...])	Effect size converter for metric involving a single group/set of scores.
effectsize.TwoSampleEffectSizeConverter([...])	Effect size converter for two-sample comparisons.
effectsize.Expression(expression[, ...])	Represent a single statistical expression.
effectsize.solve_system(system[, known_vars])	Solve and evaluate a system of SymPy equations given known inputs.
effectsize.select_expressions(target, known_vars)	Select a minimal system of expressions needed to solve for the target.
effectsize.compute_measure(measure[, data, ...])	Auto-detect and apply the right converter class.

pymare.stats: Miscellaneous statistical functions

Miscellaneous statistical functions.

Vocabulary

A group is the unit of dependence: rows sharing a label in g came from one sampling unit and are not independent of each other. That is the word used throughout for the unit itself, and it matches the g/groups arguments.

The word cluster is retained only where it is the established term for a method rather than a description of the data – “cluster-robust variance estimation”, cluster_robust_cov(), MIN_CLUSTERS_FOR_RVE, and the CR0/CR2 family. Those names tie the code to its own citations and are not renamed. A group and a cluster are the same thing; only the provenance of the word differs.

stats.weighted_least_squares(y, v, X[, ...])	Perform 2-D weighted least squares.
stats.knapp_hartung_cov_and_dof(y, v, X, ...)	Apply the Knapp-Hartung adjustment to a model-based covariance.
stats.bounded_scalar_min(f, lower, upper[, ...])	Minimize a scalar function over a bounded interval for many datasets at once.
stats.ensure_2d(arr)	Ensure the passed array has 2 dimensions.
stats.q_profile(y, v, X[, alpha, groups])	Get the CI for tau^2 via the Q-Profile method.
stats.q_gen(y, v, X, tau2[, groups])	Calculate a generalized form of Cochran's Q-statistic.
stats.log_chi2_sf(q, df)	Natural logarithm of the chi-squared upper tail, without underflowing.
stats.bonferroni(p_values)	Perform Bonferroni correction on p values.
stats.fdr(p_values[, q, method])	Perform FDR correction on p values.

pymare.datasets: Meta-analytic datasets

Open meta-analytic datasets.

datasets.michael2013()

Load a dataset of studies on the persuasive power of a brain image.

pymare.utils: Miscellaneous utility functions

Miscellaneous utility functions.

utils.get_resource_path()

Return the path to general resources, terminated with separator.