pymare.estimators.WeightedLeastSquares

class WeightedLeastSquares(tau2=0.0)

Bases: BaseEstimator

Weighted least-squares meta-regression.

Provides the weighted least-squares estimate of the fixed effects given known/assumed between-study variance tau^2, as described in Brockwell and Gordon[1]. When tau^2 = 0 (default), the model is the standard inverse-weighted fixed-effects meta-regression.

Parameters:

tau2 (float or numpy.ndarray of shape (d), optional) – Assumed/known value of tau^2. Must be >= 0. If an array, must have d elements, where d refers to the number of datasets. Default = 0.

Notes

This estimator accepts 2-D inputs for y and v–i.e., it can produce estimates simultaneously for multiple independent sets of y/v values (use the 2nd dimension for the parallel iterates). The X matrix must be identical for all iterates. If no v argument is passed to fit(), unit weights will be used, resulting in the ordinary least-squares (OLS) solution.

References

[1]

Sarah E Brockwell and Ian R Gordon. A comparison of statistical methods for meta-analysis. Statistics in medicine, 20(6):825–840, 2001. URL: https://doi.org/10.1002/sim.650, doi:10.1002/sim.650.

fit(y, X, v=None)

Fit the estimator to data.

Parameters:

• y (numpy.ndarray of shape (n, d)) – The dependent variable(s) (y).

• X (numpy.ndarray of shape (n, p)) – The independent variable(s) (X).

• v (numpy.ndarray of shape (n, d), optional) – Sampling variances. If not provided, unit weights will be used.

Return type:

WeightedLeastSquares

fit_dataset(dataset, *args, **kwargs)

Apply the current estimator to the passed Dataset container.

A convenience interface that wraps fit() and automatically aligns the variables held in a Dataset with the required arguments.

Parameters:

• dataset (Dataset) – A PyMARE Dataset instance holding the data.

• *args – Optional positional arguments to pass onto the fit() method.

• **kwargs – Optional keyword arguments to pass onto the fit() method.

get_v(dataset)

Get the variances, or an estimate thereof, from the given Dataset.

Parameters:

dataset (Dataset) – The dataset to use to retrieve/estimate v.

Returns:

2-dimensional array of variances/variance estimates.

Return type:

numpy.ndarray

Notes

This is equivalent to directly accessing dataset.v when variances are present, but affords a way of estimating v from sample size (n) for any estimator that implicitly estimates a sigma^2 parameter.

summary()

Generate a MetaRegressionResults object for the fitted estimator.

Return type:

MetaRegressionResults

Examples using pymare.estimators.WeightedLeastSquares

The Basics of Running a Meta-Analysis

The Basics of Running a Meta-Analysis

Run Estimators on a simulated dataset

Run Estimators on a simulated dataset