arXiv 13 Aug 2024 · Econometrics
arXiv:2408.06624 · PDF · DOI · OpenAlex · Extracted main text
In semi-logarithmic regressions, treatment coefficients are often interpreted as approximations of the average treatment effect (ATE) in percentage points. This paper highlights the overlooked bias of this approximation under treatment effect heterogeneity, arising from Jensen's inequality. The issue is particularly relevant for difference-in-differences designs with log-transformed outcomes and staggered treatment adoption, where treatment effects often vary across groups and periods. I propose new estimation and inference methods for the ATE in percentage points, which are applicable when treatment effects vary across and within groups. I establish the methods' large-sample properties and demonstrate their finite-sample performance through simulations, revealing substantial discrepancies between conventional and proposed measures. Two empirical applications further underscore the practical importance of these methods.
appendix boundary found by appendix_titled_section at “Online Appendix for ``Estimation and Inference on Average Treatment Effect in Percentage Points under Heterogeneity''” · 45% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Fan, Yanqin and Guerre, Emmanuel and Zhu, Dongming Partial Identification of Functionals of the Joint Distribution of “Potential Outcomes” | 0.874 | 7 | 2 | 100% |
| 2 | Chen, Jiafeng and Roth, Jonathan Logs with Zeros? Some Problems and Solutions | 0.874 | 5 | 2 | 100% |
| 3 | Sun, Liyang and Abraham, Sarah Estimating Dynamic Treatment Effects in Event Studies with Heterogeneous Treatment Effects | 0.855 | 8 | 3 | 62% |
| 4 | Callaway, Brantly and Sant'Anna, Pedro H. C Difference-in-Differences with Multiple Time Periods | 0.843 | 4 | 3 | 75% |
| 5 | Atkin, David and Khandelwal, Amit K. and Osman, Adam Exporting and Firm Performance: Evidence from a Randomized Experiment | 0.718 | 9 | 2 | 56% |
| 6 | Alsan, Marcella and Goldin, Claudia Watersheds in Child Mortality: The Role of Effective Water and Sewerage Infrastructure, 1880–1920 | 0.693 | 6 | 1 | 100% |
| 7 | Gibbons, Charles E. and Suárez Serrato, Juan Carlos and Urbancic, Mi… Broken or Fixed Effects? | 0.644 | 3 | 2 | 67% |
| 8 | Goldsmith-Pinkham, Paul and Hull, Peter and Kolesár, Michal Contamination Bias in Linear Regressions | 0.644 | 3 | 2 | 67% |
| 9 | Mullahy, John and Norton, Edward C Why Transform Y? The Pitfalls of Transformed Regressions with a Mass at Zero | 0.644 | 2 | 2 | 100% |
| 10 | de Chaisemartin, Clément and D'Haultfœuille, Xavier Two-Way Fixed Effects Estimators with Heterogeneous Treatment Effects | 0.585 | 3 | 1 | 100% |
Showing the top 10 of 31 scored citations.