Testing weak nulls in matched observational studies
We develop sensitivity analyses for weak nulls in matched observational studies while allowing unit-level treatment effects to vary. The methods may be applied to studies using any optimal without-replacement matching algorithm. In contrast to randomized experiments and to paired observational st...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023
|
| Online Access: | https://hdl.handle.net/1721.1/147999 |
| _version_ | 1826201206490398720 |
|---|---|
| author | Fogarty, Colin B |
| author2 | Massachusetts Institute of Technology. Operations Research Center |
| author_facet | Massachusetts Institute of Technology. Operations Research Center Fogarty, Colin B |
| author_sort | Fogarty, Colin B |
| collection | MIT |
| description | We develop sensitivity analyses for weak nulls in matched observational
studies while allowing unit-level treatment effects to vary. The methods may be
applied to studies using any optimal without-replacement matching algorithm. In
contrast to randomized experiments and to paired observational studies, we show
for general matched designs that over a large class of test statistics, any
valid sensitivity analysis for the entirety of the weak null must be
unnecessarily conservative if Fisher's sharp null of no treatment effect for
any individual also holds. We present a sensitivity analysis valid for the weak
null, and illustrate why it is generally conservative if the sharp null holds
through new connections to inverse probability weighted estimators. An
alternative procedure is presented that is asymptotically sharp if treatment
effects are constant, and that is valid for the weak null under additional
restrictions which may be deemed benign by practitioners. Simulations
demonstrate that this alternative procedure results in a valid sensitivity
analysis for the weak null hypothesis under a host of reasonable
data-generating processes. The procedures allow practitioners to assess
robustness of estimated sample average treatment effects to hidden bias while
allowing for unspecified effect heterogeneity in matched observational studies. |
| first_indexed | 2024-09-23T11:47:52Z |
| format | Article |
| id | mit-1721.1/147999 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T11:47:52Z |
| publishDate | 2023 |
| publisher | Wiley |
| record_format | dspace |
| spelling | mit-1721.1/1479992023-02-11T03:30:45Z Testing weak nulls in matched observational studies Fogarty, Colin B Massachusetts Institute of Technology. Operations Research Center We develop sensitivity analyses for weak nulls in matched observational studies while allowing unit-level treatment effects to vary. The methods may be applied to studies using any optimal without-replacement matching algorithm. In contrast to randomized experiments and to paired observational studies, we show for general matched designs that over a large class of test statistics, any valid sensitivity analysis for the entirety of the weak null must be unnecessarily conservative if Fisher's sharp null of no treatment effect for any individual also holds. We present a sensitivity analysis valid for the weak null, and illustrate why it is generally conservative if the sharp null holds through new connections to inverse probability weighted estimators. An alternative procedure is presented that is asymptotically sharp if treatment effects are constant, and that is valid for the weak null under additional restrictions which may be deemed benign by practitioners. Simulations demonstrate that this alternative procedure results in a valid sensitivity analysis for the weak null hypothesis under a host of reasonable data-generating processes. The procedures allow practitioners to assess robustness of estimated sample average treatment effects to hidden bias while allowing for unspecified effect heterogeneity in matched observational studies. 2023-02-10T13:39:12Z 2023-02-10T13:39:12Z 2019-08-20 2023-02-10T13:31:41Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/147999 Fogarty, Colin B. 2019. "Testing weak nulls in matched observational studies." Biometrics. en 10.1111/biom.13741 Biometrics Creative Commons Attribution 4.0 International license https://creativecommons.org/licenses/by/4.0/ application/pdf Wiley Wiley |
| spellingShingle | Fogarty, Colin B Testing weak nulls in matched observational studies |
| title | Testing weak nulls in matched observational studies |
| title_full | Testing weak nulls in matched observational studies |
| title_fullStr | Testing weak nulls in matched observational studies |
| title_full_unstemmed | Testing weak nulls in matched observational studies |
| title_short | Testing weak nulls in matched observational studies |
| title_sort | testing weak nulls in matched observational studies |
| url | https://hdl.handle.net/1721.1/147999 |
| work_keys_str_mv | AT fogartycolinb testingweaknullsinmatchedobservationalstudies |