From LATE to MTE: Alternative methods for the evaluation of policy interventions
Open accessThe DOI matches its Crossref record; 8 of the 8 references it cites were found in a registry or an index.
Abstract
This paper provides an introduction into the estimation of marginal treatment effects (MTE). Compared to the existing surveys on the subject, our paper is less technical and speaks to the applied economist with a solid basic understanding of econometric techniques who would like to use MTE estimation. Our framework of analysis is a generalized Roy model based on the potential outcomes framework, within which we define different treatment effects of interest, and review the well-known case of IV estimation with a discrete instrument resulting in a local average treatment effect (LATE). Turning to IV estimation with a continuous instrument, we demonstrate that the 2SLS estimator may be viewed as a weighted average of LATEs and discuss MTE estimation as an alternative and more informative way of exploiting a continuous instrument. We clarify the assumptions underlying the MTE framework, its relation to the correlated random coefficients model, and illustrate how the MTE estimation is implemented in practice.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Estimating causal effects of treatments in randomized and nonrandomized studies
Rubin, D. B. (1974) · Journal of Educational Psychology · doi:10.1037/h0037350 · Cited by 9,436 · cited 2×
“We model these potential outcomes as Y0i = μ0(Xi) + U0i (1) Y1i = μ1(Xi) + U1i (2) 4 The potential outcome model, often also referred to as the “Rubin causal model” is a building block for the literature on causal inference and goes back to Rubin (1974) and Holland (1986).”
BackgroundStatistics and Causal Inference
Holland, P. W. (1986) · Journal of the American Statistical Association · doi:10.1080/01621459.1986.10478354 · Cited by 5,293 · cited 1×
“We model these potential outcomes as Y0i = μ0(Xi) + U0i (1) Y1i = μ1(Xi) + U1i (2) 4 The potential outcome model, often also referred to as the “Rubin causal model” is a building block for the literature on causal inference and goes back to Rubin (1974) and Holland (1986).”
BackgroundIdentification of Causal Effects Using Instrumental Variables
Angrist, J. D., Imbens, G. W., & Rubin, D. B. (1996) · Journal of the American Statistical Association · doi:10.1080/01621459.1996.10476902 · Cited by 4,230
Identification and estimation of local average treatment effects
Imbens, G. W., & Angrist, J. D. (1994) · Econometrica · doi:10.2307/2951620 · Cited by 4,101
Chetty, R., Hendren, N., & Katz, L. F. (2016) · American Economic Review · doi:10.1257/aer.20150572 · Cited by 2,315
Structural Equations, Treatment Effects, and Econometric Policy Evaluation1
Heckman, J. J., & Vytlacil, E. (2005) · Econometrica · doi:10.1111/j.1468-0262.2005.00594.x · Cited by 1,220
Estimating Marginal Returns to Education
Carneiro, P., Heckman, J. J., & Vytlacil, E. J. (2011) · American Economic Review · doi:10.1257/aer.101.6.2754 · Cited by 612 · cited 21×
“As in Carneiro et al. (2011), the linear IV estimate turns out to be similar in magnitude to ATE and masks important heterogeneity in returns.”
BackgroundMethodResultInfluentialWho benefits from universal child care? Estimating marginal returns to early child care attendance
Cornelissen, T., Dustmann, C., Raute, A., & Schönberg, U. (2018) · Journal of Political Economy · doi:10.1086/699979 · Cited by 315
Cited by
Using and Interpreting Fixed Effects Models
BREUER, M., & DEHAAN, E. (2024) · Journal of Accounting Research · doi:10.1111/1475-679x.12559 · Cited by 526
Racial Bias in Bail Decisions*
Arnold, D., Dobbie, W., & Yang, C. S. (2018) · The Quarterly Journal of Economics · doi:10.1093/qje/qjy012 · Cited by 387
Who benefits from universal child care? Estimating marginal returns to early child care attendance
Cornelissen, T., Dustmann, C., Raute, A., & Schönberg, U. (2018) · Journal of Political Economy · doi:10.1086/699979 · Cited by 315
“33 As explained in detail in Cornelissen et al. (2016), t he 2SLS estimator may be viewed as a weighted average of Local Average Treatment Effects across rr’ pairs, when group indicator dummies R i are used as instruments. The overall IV estimate is therefore represen tative for compliers at all values of the instrume…”
InfluentialWomen's Empowerment in Action: Evidence from a Randomized Control Trial in Africa
Bandiera, O., Buehren, N., Burgess, R., Goldstein, M., Gulesci, S., Rasul, I., & Sulaiman, M. (2020) · American Economic Journal: Applied Economics · doi:10.1257/app.20170416 · Cited by 311
Endogeneity in survey research
Sande, J. B., & Ghosh, M. (2018) · International Journal of Research in Marketing · doi:10.1016/j.ijresmar.2018.01.005 · Cited by 264
Metadata
- Authors
- Cornelissen, T., Dustmann, C., Raute, A., & Schönberg, U.
- Venue
- Labour Economics
- Publisher
- Elsevier BV
- Published
- 2016-08
- Volume
- 41
- Pages
- 47-60
- ISSN
- 0927-5371 (ISSN-L)
- OpenAlex
- W3121219797
- Cited by
- 143
- References
- 8
- FWCI
- 9.41
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.