Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
110 tokens/sec
GPT-4o
56 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Estimating Causal Effects with Double Machine Learning -- A Method Evaluation (2403.14385v2)

Published 21 Mar 2024 in stat.ML, cs.LG, econ.EM, and stat.ME

Abstract: The estimation of causal effects with observational data continues to be a very active research area. In recent years, researchers have developed new frameworks which use machine learning to relax classical assumptions necessary for the estimation of causal effects. In this paper, we review one of the most prominent methods - "double/debiased machine learning" (DML) - and empirically evaluate it by comparing its performance on simulated data relative to more traditional statistical methods, before applying it to real-world data. Our findings indicate that the application of a suitably flexible machine learning algorithm within DML improves the adjustment for various nonlinear confounding relationships. This advantage enables a departure from traditional functional form assumptions typically necessary in causal effect estimation. However, we demonstrate that the method continues to critically depend on standard assumptions about causal structure and identification. When estimating the effects of air pollution on housing prices in our application, we find that DML estimates are consistently larger than estimates of less flexible methods. From our overall results, we provide actionable recommendations for specific choices researchers must make when applying DML in practice.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (85)
  1. Country of origin: A competitive advantage? International Journal of Research in Marketing, 16(4):255–267.
  2. Pricing for Heterogeneous Products: Analytics for Ticket Reselling. Manufacturing & Service Operations Management, 25(2):409–426.
  3. Mostly Harmless Econometrics: An Empiricist’s Companion. Princeton University Press, Princeton, 1 edition.
  4. Athey, S. (2019). The Impact of Machine Learning on Economics. In Agrawal, A., Gans, J., and Goldfarb, A., editors, The Economics of Artificial Intelligence: An Agenda, pages 507–552. University of Chicago Press, Chicago.
  5. The State of Applied Econometrics: Causality and Policy Evaluation. The Journal of Economic Perspectives, 31(2):3–32.
  6. Long-term effects from early exposure to research: Evidence from the NIH “Yellow Berets”. Research Policy, 50(9):104332.
  7. Beach, E. F. (1949). The Use of Polynomials to Represent Cost Functions. The Review of Economic Studies, 16(3):158–169.
  8. Program Evaluation and Causal Inference with High-Dimensional Data. Econometrica, 85(1):233–298.
  9. High-Dimensional Methods and Inference on Structural and Treatment Effects. Journal of Economic Perspectives, 28(2):29–50.
  10. Inference in High-Dimensional Panel Models With an Application to Gun Control. Journal of Business & Economic Statistics, 34(4):590–605.
  11. New Empirical Generalizations on the Determinants of Price Elasticity. Journal of Marketing Research, 42(2):141–156.
  12. Prosocial behavior in emergencies: Evidence from blood donors recruitment and retention during the COVID-19 pandemic. Social Science & Medicine, 314:115438.
  13. Evaluating (weighted) dynamic treatment effects by double machine learning. Econometrics Journal, 25(3):628–648.
  14. Card, D. (1999). The Causal Effect of Education on Earnings. In Ashenfelter, O. C. and Card, D., editors, Handbook of Labor Economics, volume 3, pages 1801–1863. Elsevier.
  15. Behind the screen: Understanding national support for a foreign investment screening mechanism in the European Union. The Review of International Organizations, 17(3):513–541.
  16. Chang, N.-C. (2020). Double/debiased machine learning for difference-in-differences models. The Econometrics Journal, 23(2):177–191.
  17. Debiased/Double Machine Learning for Instrumental Variable Quantile Regressions. Econometrics, 9(2):15.
  18. xgboost: Extreme Gradient Boosting.
  19. Double/debiased machine learning for treatment and structural parameters. The Econometrics Journal, 21(1):C1–C68.
  20. Causal impact of masks, policies, behavior on early covid-19 pandemic in the U.S. Journal of Econometrics, 220(1):23–62.
  21. Multiway Cluster Robust Double/Debiased Machine Learning. Journal of Business & Economic Statistics, 40(3):1046–1056.
  22. A Crash Course in Good and Bad Controls. Sociological Methods & Research, pages 1–34.
  23. Smoking and lung cancer: recent evidence and a discussion of some questions*. International Journal of Epidemiology, 38(5):1175–1191.
  24. Youth well-being predicts later academic success. Scientific Reports, 12(1):2134.
  25. doParallel: Foreach Parallel Adaptor for the ’parallel’ Package.
  26. foreach: Provides Foreach Looping Construct.
  27. Corruption red flags in public procurement: new evidence from Italian calls for tenders. EPJ Data Science, 11(1):1–38.
  28. Diaz, I. (2020). Machine learning in the estimation of causal effects: targeted minimum loss-based estimation and double/debiased machine learning. Biostatistics, 21(2):353–358.
  29. Adherence, Persistence, Readmissions, and Costs in Medicaid Members with Schizophrenia or Schizoaffective Disorder Initiating Paliperidone Palmitate Versus Switching Oral Antipsychotics: A Real-World Retrospective Investigation. Advances in Therapy, 40(1):349–366.
  30. Monopsony in Online Labor Markets. American Economic Review: Insights, 2(1):33–46.
  31. Estimating Marketing Component Effects: Double Machine Learning from Targeted Digital Promotions. Marketing Science, 42(4):704–728.
  32. Causal mediation analysis with double machine learning. The Econometrics Journal, 25(2):277–300.
  33. Using Double Machine Learning to Understand Nonresponse in the Recruitment of a Mixed-Mode Online Panel. Social Science Computer Review, 41(2):461–481.
  34. Gaujoux, R. (2023). doRNG: Generic Reproducible Parallel Backend for ’foreach’ Loops.
  35. Goller, D. (2023). Analysing a built-in advantage in asymmetric darts contests using causal machine learning. Annals of Operations Research, 325:649–679.
  36. Close Enough? A Large-Scale Exploration of Non-Experimental Approaches to Advertising Measurement. Marketing Science, 42(4):768–793.
  37. Hansen, D. (2020). The effectiveness of fiscal institutions: International financial flogging or domestic constraint? European Journal of Political Economy, 63:101879.
  38. Hedonic housing prices and the demand for clean air. Journal of Environmental Economics and Management, 5(1):81–102.
  39. The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition. Springer Science & Business Media, 2 edition.
  40. Causal Inference: What If. Chapman & Hall/CRC.
  41. Interdependence and the cost of uncoordinated responses to COVID-19. Proceedings of the National Academy of Sciences, 117(33):19837–19843.
  42. Business analytics meets artificial intelligence: Assessing the demand effects of discounts on Swiss train tickets. Transportation Research Part B: Methodological, 163:22–39.
  43. Imbens, G. W. (2004). Nonparametric Estimation of Average Treatment Effects Under Exogeneity: A Review. The Review of Economics and Statistics, 86(1):4–29.
  44. Imbens, G. W. (2020). Potential Outcome and Directed Acyclic Graph Approaches to Causality: Relevance for Empirical Practice in Economics. Journal of Economic Literature, 58(4):1129–1179.
  45. Identification and Estimation of Local Average Treatment Effects. Econometrica, 62(2):467–475.
  46. Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction. Cambridge University Press, Cambridge.
  47. Recent Developments in the Econometrics of Program Evaluation. Journal of Economic Literature, 47(1):5–86.
  48. An Introduction to Statistical Learning: with Applications in R. Springer, New York, 2 edition.
  49. Knaus, M. C. (2021). A double machine learning approach to estimate the effects of musical practice on student’s skills. Journal of the Royal Statistical Society: Series A (Statistics in Society), 184(1):282–300.
  50. Knaus, M. C. (2022). Double machine learning-based programme evaluation under unconfoundedness. The Econometrics Journal, 25(3):602–627.
  51. Liquidity costs on intraday power markets: Continuous trading versus auctions. Energy Policy, 154:112299.
  52. Targeted Maximum Likelihood Learning. The International Journal of Biostatistics, 2(1):1–38.
  53. Classification and Regression by randomForest. R News, 2(3):18–22.
  54. Double/debiased machine learning for logistic partially linear model. The Econometrics Journal, 24(3):559–588.
  55. External control arm analysis: an evaluation of propensity score approaches, G-computation, and doubly debiased machine learning. BMC Medical Research Methodology, 22(1):335.
  56. Lundberg, I. (2022). The Gap-Closing Estimand: A Causal Approach to Study Interventions That Close Disparities Across Social Categories. Sociological Methods & Research, pages 1–64.
  57. Practical variable selection for generalized additive models. Computational Statistics & Data Analysis, 55(7):2372–2387.
  58. Estimating treatment effects with machine learning. Health Services Research, 54(6):1273–1282.
  59. Machine Learning: An Applied Econometric Approach. Journal of Economic Perspectives, 31(2):87–106.
  60. Nelson, J. P. (1978). Residential choice, hedonic prices, and the demand for urban air quality. Journal of Urban Economics, 5(3):357–369.
  61. The association between experiencing homelessness in childhood or youth and adult housing stability in Housing First. BMC Psychiatry, 21(1):138.
  62. Pearl, J. (2009). Causality. Cambridge University Press, Cambridge, 2 edition.
  63. Pearl, J. (2016). Causal Inference in Statistics: A Primer. Wiley, Chichester, West Sussex, 1 edition.
  64. The Book of Why: The New Science of Cause and Effect. Penguin, London, 1 edition.
  65. Powell, J. L. (1994). Estimation of semiparametric models. In Engle, R. F. and McFadden, D. L., editors, Handbook of Econometrics, volume 4, pages 2443–2521. Elsevier.
  66. Statistical and machine learning methods for evaluating trends in air quality under changing meteorological conditions. Atmospheric Chemistry and Physics, 22(16):10551–10566.
  67. R Core Team (2023). R: A Language and Environment for Statistical Computing.
  68. Semiparametric Efficiency in Multivariate Regression Models with Missing Data. Journal of the American Statistical Association, 90(429):122–129.
  69. Estimation of Regression Coefficients When Some Regressors Are Not Always Observed. Journal of the American Statistical Association, 89(427):846–866.
  70. Robinson, P. M. (1988). Root-N-Consistent Semiparametric Regression. Econometrica, 56(4):931–954.
  71. Rosen, S. (1974). Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition. Journal of Political Economy, 82(1):34–55.
  72. The Central Role of the Propensity Score in Observational Studies for Causal Effects. Biometrika, 70(1):41–55.
  73. Debiased machine learning of conditional average treatment effects and other causal functions. The Econometrics Journal, 24(2):264–289.
  74. Shmueli, G. (2010). To Explain or to Predict? Statistical Science, 25(3):289–310.
  75. Regularization Paths for Cox’s Proportional Hazards Model via Coordinate Descent. Journal of Statistical Software, 39(5):1–13.
  76. Child stature, maternal education, and early childhood development in Nigeria. PLOS ONE, 16(12):1–17.
  77. StataCorp (2019). Stata Statistical Software: Release 16.
  78. Assumption-Lean Cox Regression. Journal of the American Statistical Association, pages 1–10.
  79. Modern Applied Statistics with S. Springer, New York, fourth edition.
  80. Wood, S. N. (2017). Generalized Additive Models: An Introduction with R. Chapman and Hall/CRC, 2 edition.
  81. Wooldridge, J. (2012). Introductory Econometrics: A Modern Approach. Cengage Learning, Inc, Mason, OH, 5 edition.
  82. Wooldridge, J. M. (2010). Econometric Analysis of Cross Section and Panel Data. MIT Press, Cambridge, MA, USA, 2 edition.
  83. Is the younger generation a driving force toward achieving the sustainable development goals? Survey experiments. Journal of Cleaner Production, 292:125932.
  84. Double machine learning with gradient boosting and its application to the Big N audit quality effect. Journal of Econometrics, 216(1):268–283.
  85. Machine Learning for Causal Inference: On the Use of Cross-fit Estimators. Epidemiology, 32(3):393.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (3)
  1. Jonathan Fuhr (2 papers)
  2. Philipp Berens (27 papers)
  3. Dominik Papies (3 papers)
Citations (2)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Youtube Logo Streamline Icon: https://streamlinehq.com

YouTube