A Combinatorial Central Limit Theorem for Stratified Randomization
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910410661167104 |
|---|---|
| author | Tuvaandorj, Purevdorj |
| author_facet | Tuvaandorj, Purevdorj |
| contents | This paper establishes a combinatorial central limit theorem for stratified randomization, which holds under a Lindeberg-type condition. The theorem allows for an arbitrary number or sizes of strata, with the sole requirement being that each stratum contains at least two units. This flexibility accommodates both a growing number of large and small strata simultaneously, while imposing minimal conditions. We then apply this result to derive the asymptotic distributions of two test statistics proposed for instrumental variables settings in the presence of potentially many strata of unrestricted sizes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_14764 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Combinatorial Central Limit Theorem for Stratified Randomization Tuvaandorj, Purevdorj Statistics Theory Econometrics This paper establishes a combinatorial central limit theorem for stratified randomization, which holds under a Lindeberg-type condition. The theorem allows for an arbitrary number or sizes of strata, with the sole requirement being that each stratum contains at least two units. This flexibility accommodates both a growing number of large and small strata simultaneously, while imposing minimal conditions. We then apply this result to derive the asymptotic distributions of two test statistics proposed for instrumental variables settings in the presence of potentially many strata of unrestricted sizes. |
| title | A Combinatorial Central Limit Theorem for Stratified Randomization |
| topic | Statistics Theory Econometrics |
| url | https://arxiv.org/abs/2402.14764 |