Abstract
The publisher regrets some errors in the published paper. The publisher would like to apologise for any inconvenience caused. Updated version is as below: Page 3, first formula: [Formula presented] Page 3, second formula: [Formula presented] Page 3, third formula (overlap): [Formula presented] Page 3 section 2.2, derivations: (1) Because [Formula presented] (i.e., observed outcomes) is [Formula presented] when [Formula presented], we have [Formula presented].(2) By iterated expectations, we have [Formula presented].(3) By unconfoundedness, we have [Formula presented].(4) And because [Formula presented], the right-hand side cancels to give [Formula presented], from which we have [Formula presented] as desired (Williamson et al., 2012). Similarly, [Formula presented]. Page 3 section 2.2, ATT estimation: [Formula presented] Page 3 section 2.2 last paragraph, King and Nielsen (2019): Reference: King, G., Nielsen, R., 2019. Why Propensity Scores Should Not Be Used for Matching. Political Analysis, 27(4), 435–454. https://doi.org/10.1017/pan.2019.11.
Original language | English |
---|---|
Article number | 106257 |
Pages (from-to) | 106257 |
Number of pages | 1 |
Journal | Accident Analysis and Prevention |
Volume | 159 |
DOIs |
|
Publication status | Published - 1 Sept 2021 |
ASJC Scopus subject areas
- Human Factors and Ergonomics
- Safety, Risk, Reliability and Quality
- Public Health, Environmental and Occupational Health