General Forms of Finite Population Central Limit Theorems with Applications to Causal Inference

By Xinran Li, Peng Ding. 2016

210 citations recorded by the source corpus.

ABSTRACT Frequentists’ inference often delivers point estimators associated with confidence intervals or sets for parameters of interest. Constructing the confidence intervals or sets requires understanding the sampling distributions of the point estimators, which, in many but not all cases, are related to asymptotic Normal distributions ensured by central limit theorems. Although previous literature has established various forms of central limit theorems for statistical inference in super population models, we still need general and convenient forms of central limit theorems for some randomization-based causal analyses of experimental data, where the parameters of interests are functions of a finite population and randomness comes solely from the treatment assignment. We use central limit theorems for sample surveys and rank statistics to establish general forms of the finite population central limit theorems that are particularly useful for proving asymptotic distributions of randomization tests under the sharp null hypothesis of zero individual causal effects, and for obtaining the asymptotic repeated sampling distributions of the causal effect estimators. The new central limit theorems hold for general experimental designs with multiple treatment levels, multiple treatment factors and vector outcomes, and are immediately applicable for studying the asymptotic properties of many methods in causal inference, including instrumental variable, regression adjustment, rerandomization, cluster-randomized experiments, and so on. Previously, the asymptotic properties of these problems are often based on heuristic arguments, which in fact rely on general forms of finite population central limit theorems that have not been established before. Our new theorems fill this gap by providing more solid theoretical foundation for asymptotic randomization-based causal inference. Supplementary materials for this article are available online.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/ad552d409d3aa0ba5dd276f7914426b648c826a6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

Rerandomization and regression adjustment (2019, methodology): “The finite population central limit theorem of Li and Ding (2017) ensures that n1/2(τ̂−τ, τ̂ ′ x, τ̂ ′ w) is asymptotically Gaussian with mean 0 and covariance matrix V under the CRE and Condition 1.”
Randomization Tests for Weak Null Hypotheses in Randomized Experiments (2018, background): “Neuhaus (1993), Janssen (1997), Janssen (1999), Janssen and Pauls (2003) and Chung and Romano (2013) conducted permutation tests with studentization.”
Panel experiments and dynamic causal effects: A finite population perspective (2020, background): “As discussed in Li and Ding (2017), this assumption can often be replaced by a finite-population analogue of the Lindeberg condition in analyses of cross-sectional, randomized experiments.”
Decomposing Treatment Effect Variation (2016, methodology): “A recent review is Li and Ding (2017).”
Regression-adjusted average treatment effect estimates in stratified randomized experiments (2019, methodology): “Finite population central limit theorem plays an essential role for randomization-based asymptotic inference [20, 29, 33].”
The Generalized Oaxaca-Blinder Estimator (2020, methodology): “The tools in Li and Ding (2020) can be used to study the asymptotic distribution of an asymptotically linear statistic under Mahalanobis rerandomization, and the delta method can be used to derive the asymptotic distribution of other smooth functions of the imputed means.”