Subsampling Under Two-way Clustering with Serial Correlation
TLDR
This paper validates subsampling for inference in two-way clustered panels with serial correlation, introducing methods for non-Gaussian limits.
Key contributions
- Validates subsampling for inference under two-way clustered panels with serially correlated time effects.
- Introduces a quantile method enabling inference under non-Gaussian asymptotic limits, a novel contribution.
- Presents a variance method with data-driven bandwidth and bias-correction for Gaussian limits.
- Demonstrates robust coverage in simulations, addressing a critical gap in existing methods.
Why it matters
This work is crucial as it provides the first method for valid inference under non-Gaussian asymptotics in two-way clustered panels with serial correlation, a common challenge in econometrics. It expands the applicability of subsampling, offering more robust statistical tools for complex panel data analysis.
Original Abstract
We prove the validity of using subsampling method for inference under a two-way clustered panel in which the time effects are serially correlated. Subsamples should be drawn without replacement from randomly partitioned individual index set and consecutive blocks of time effects. We present two subsampling inference methods: estimating the quantiles directly and constructing the confidence interval by first estimating the asymptotic variance. The quantile method is very adaptive, allowing for non-Gaussian limit which invalidates all existing methods in two-way clustering with serial correlation. Although the variance method only works under Gaussian limit, it comes with a data-driven bandwidth selection algorithm and a bias-correction under suitable estimators. Monte Carlo simulations demonstrate our methods exhibiting the desired coverage level in the finite sample except when the serial correlation is extremely strong. This paper is the first one that allows for inference on non-Gaussian asymptotics under two-way clustering with serial correlation.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.