Robust inference on average treatment effects with possibly more covariates than observations

Farrell, Max H.

doi:10.1016/j.jeconom.2015.06.017

Cited by 218 publications

(158 citation statements)

References 88 publications

Supporting

Mentioning

151

Contrasting

Order By: Relevance

“…Loosely speaking, as long as the true conditional mean function can be closely approximated by a model using fewer covariates than observations, the LASSO can serve as a highly accurate model selection algorithm. Furthermore, it is straightforward to construct accurate confidence intervals for causal effects of interest after performing model selection with the LASSO Farrell 2015).…”

Section: Estimationmentioning

confidence: 99%

“…In estimating the ATT, we follow a procedure detailed by Farrell (2015) that combines "doubly robust" treatment effect estimation with model selection using the LASSO algorithm. We use the LASSO to select from a set of variables that are potentially correlated with our outcomes of interest as well as RBD participation under a unitary agricultural household model.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Are matching funds for smallholder irrigation money well spent?

Mullally

Chakravarty

2018

Food Policy

View full text Add to dashboard Cite

Groundwater irrigation can dramatically affect agricultural production and productivity. Despite its potential as an agricultural development tool, little credible evidence exists for the impacts of groundwater development on smallholder agriculture. We add to the evidence on the benefits of irrigation investments for small producers by evaluating the Rural Business Development (RBD) program of the Millennium Challenge Corporation in Nicaragua for small plantain producers. The RBD program offered matching funds covering up to 30% of the cost of two years of inputs, extension services, and dieselpowered micro-sprinkler irrigation for individual farms. In order to estimate the average impact of the RBD program on its beneficiaries, we combine model selection via the LASSO with doubly robust treatment effect estimation as applied to two years of panel data for 146 producers. We find that the program had substantial impacts on plantain revenue and production, while achieving nearly complete irrigation coverage of plantain land among beneficiaries. We find no discernible impact on household expenditure. JEL Codes: O13, O22, Q12

show abstract

Section: Estimationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Are matching funds for smallholder irrigation money well spent?

Mullally

Chakravarty

2018

Food Policy

View full text Add to dashboard Cite

show abstract

“…In addition, as one of their remarkable contributions, Belloni et al (2013) demonstrate that post-selection inference with their Lasso procedures yields a uniformly valid inference procedure for ATE and ATT. See also Farrell (2013), who derives uniformly valid inference procedures in a similar setup.…”

Section: Related Literaturementioning

confidence: 99%

“…12 We recommend using the uniform prior, unless the user has a strong prior opinion about the value of δ for the covariates. In our Monte Carlo studies and empirical application, we examine performance of the BayesLE-averaging estimator with the uniform prior.…”

Section: At T Smentioning

confidence: 99%

Model averaging in semiparametric estimation of treatment effects

Kitagawa

Muris²

2015

View full text Add to dashboard Cite

In the practice of program evaluation, choosing the covariates and the functional form of the propensity score is an important choice that the researchers make when estimating treatment effects. This paper proposes a data-driven way of averaging the estimators over the candidate specifications in order to resolve the issue of specification uncertainty in the propensity score weighting estimation of the average treatment effects for treated (ATT). The proposed averaging procedures aim to minimize the estimated mean squared error (MSE) of the ATT estimator in a local asymptotic framework. We formulate model averaging as a statistical decision problem in a limit experiment, and derive an averaging scheme that is Bayes optimal with respect to a given prior for the localization parameters. Analytical comparisons of the Bayes asymptotic MSE show that the averaging estimator outperforms post model selection estimators and the estimators in any of the candidate models. Our Monte Carlo studies confirm these theoretical results and illustrate the size of the MSE gains from averaging. We apply the averaging procedure to evaluate the effect of the labor market program analyzed in LaLonde (1986).Keywords: Treatment effects, Propensity score, Model averaging, Limit experiment.JEL Classification: C13, C21, C52. * We thank Debopam Bhattacharya, Irene Botosaru, Xiaohong Chen, Christian Hansen, Yuichi Kitamura, Frank Kleibergen, Michael Lechner, Simon Lee, Richard Smith, and Frank Windmeijer for valuable comments and discussions. We thank an associate editor and three referees for their thorough reading and constructive comments on an earlier draft. We also thank the seminar participants at AMES 2013, University of Bristol, University of British Columbia, Brown University, the Cemmap/PEPA workshop on Program Evaluation, University of Groningen, University of Sankt Gallen, Tilburg University, Toulouse School of Economics, and Yale Econometrics Lunch for their helpful comments. All remaining errors are ours. Financial support from the ESRC through the ESRC Centre for Microdata Methods and Practice (CeMMAP) (grant number RES-589-28-0001) is gratefully acknowledged.

show abstract

“…15 Note that this condition is stronger than necessary but convenient; see, e.g., Farrell (2013). We briefly explore this in the simulation example in Section 5 where we consider a design where Condition ASM holds only in one equation and show that our procedure still yields good results in that setting.…”

Section: Selection Of Control Variablesmentioning

confidence: 99%

Inference in high dimensional panel models with an application to gun control

Belloni¹,

Chernozhukov²,

Hansen³

et al. 2014

View full text Add to dashboard Cite

Abstract. We consider estimation and inference in panel data models with additive unobserved individual specific heterogeneity in a high dimensional setting. The setting allows the number of time varying regressors to be larger than the sample size. To make informative estimation and inference feasible, we require that the overall contribution of the time varying variables after eliminating the individual specific heterogeneity can be captured by a relatively small number of the available variables whose identities are unknown. This restriction allows the problem of estimation to proceed as a variable selection problem. Importantly, we treat the individual specific heterogeneity as fixed effects which allows this heterogeneity to be related to the observed time varying variables in an unspecified way and allows that this heterogeneity may be non-zero for all individuals. Within this framework, we provide procedures that give uniformly valid inference over a fixed subset of parameters in the canonical linear fixed effects model and over coefficients on a fixed vector of endogenous variables in panel data instrumental variables models with fixed effects and many instruments. An input to developing the properties of our proposed procedures is the use of a variant of the Lasso estimator that allows for a grouped data structure where data across groups are independent and dependence within groups is unrestricted. We provide formal conditions within this structure under which the proposed Lasso variant selects a sparse model with good approximation properties. We present simulation results in support of the theoretical developments and illustrate the use of the methods in an application aimed at estimating the effect of gun prevalence on crime rates.Key Words: panel data, fixed effects, partially linear model, instrumental variables, high dimensional-sparse regression, inference under imperfect model selection, uniformly valid inference after model selection, clustered standard errors

show abstract

Robust inference on average treatment effects with possibly more covariates than observations

Cited by 218 publications

References 88 publications

Are matching funds for smallholder irrigation money well spent?

Are matching funds for smallholder irrigation money well spent?

Model averaging in semiparametric estimation of treatment effects

Inference in high dimensional panel models with an application to gun control

Contact Info

Product

Resources

About