Magging: maximin aggregation for inhomogeneous large-scale data

TL;DR

Magging introduces a maximin aggregation method designed to effectively analyze large-scale, inhomogeneous data by focusing on effects common across all data subsets, improving interpretability and prediction.

Contribution

The paper proposes a novel maximin aggregation technique that enhances analysis of inhomogeneous large-scale data, addressing limitations of traditional averaging methods.

Findings

Maximin aggregation identifies effects consistent across all data subsets.

The method improves prediction accuracy over pooled effects.

It effectively handles outliers and time-varying effects.

Abstract

Large-scale data analysis poses both statistical and computational problems which need to be addressed simultaneously. A solution is often straightforward if the data are homogeneous: one can use classical ideas of subsampling and mean aggregation to get a computationally efficient solution with acceptable statistical accuracy, where the aggregation step simply averages the results obtained on distinct subsets of the data. However, if the data exhibit inhomogeneities (and typically they do), the same approach will be inadequate, as it will be unduly influenced by effects that are not persistent across all the data due to, for example, outliers or time-varying effects. We show that a tweak to the aggregation step can produce an estimator of effects which are common to all data, and hence interesting for interpretation and often leading to better prediction than pooled effects.