Difference-in-Differences with a Continuous Treatment
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2021
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909979220377600 |
|---|---|
| author | Callaway, Brantly Goodman-Bacon, Andrew Sant'Anna, Pedro H. C. |
| author_facet | Callaway, Brantly Goodman-Bacon, Andrew Sant'Anna, Pedro H. C. |
| contents | This paper analyzes difference-in-differences designs with a continuous treatment. We show that treatment-on-the-treated-type parameters are identified under a parallel trends assumption analogous to the binary treatment case. However, comparing these parameters across treatments is challenging because parallel trends does not rule out selection bias. We discuss alternative, typically stronger, assumptions that eliminate selection bias. We further show that popular two-way fixed effects estimands admit multiple interpretations, depending on the underlying causal building block, all having important limitations as meaningful summaries of treatment effects. Finally, we introduce alternative estimation procedures that avoid these drawbacks and demonstrate them in an empirical application. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2107_02637 |
| institution | arXiv |
| publishDate | 2021 |
| record_format | arxiv |
| spellingShingle | Difference-in-Differences with a Continuous Treatment Callaway, Brantly Goodman-Bacon, Andrew Sant'Anna, Pedro H. C. Econometrics This paper analyzes difference-in-differences designs with a continuous treatment. We show that treatment-on-the-treated-type parameters are identified under a parallel trends assumption analogous to the binary treatment case. However, comparing these parameters across treatments is challenging because parallel trends does not rule out selection bias. We discuss alternative, typically stronger, assumptions that eliminate selection bias. We further show that popular two-way fixed effects estimands admit multiple interpretations, depending on the underlying causal building block, all having important limitations as meaningful summaries of treatment effects. Finally, we introduce alternative estimation procedures that avoid these drawbacks and demonstrate them in an empirical application. |
| title | Difference-in-Differences with a Continuous Treatment |
| topic | Econometrics |
| url | https://arxiv.org/abs/2107.02637 |