1932

Abstract

This article reviews recent endeavors to incorporate big data and machine learning techniques into energy and environmental economics research. We find that novel datasets, from high frequency smart meter data to satellite images and social media data, are already used by researchers. At the same time most of the analyses rely on traditional econometric techniques. Nevertheless, we find applications of machine learning models that address the high dimensionality of the data and seek out new and better strategies for estimating heterogenous treatment effects. We provide an introduction to the main themes in machine learning, which are likely to be of use to economists in energy and environmental economics, and illustrate them using a real data example derived from an energy efficiency program evaluation. We provide the data and code in order to stimulate further research in this area.

Loading

Article metrics loading...

/content/journals/10.1146/annurev-resource-100920-034117
2021-10-05
2024-03-29
Loading full text...

Full text loading...

/deliver/fulltext/resource/13/1/annurev-resource-100920-034117.html?itemId=/content/journals/10.1146/annurev-resource-100920-034117&mimeType=html&fmt=ahah

Literature Cited

  1. Albanesi S, De Giorgi G, Nosal J. 2017. Credit growth and the financial crisis: a new narrative NBER Work. Pap. 23740
  2. Allcott H. 2011. Social norms and energy conservation. J. Public Econ. 95:9–101082–95
    [Google Scholar]
  3. Athey S, Imbens G 2016. Recursive partitioning for heterogeneous causal effects. PNAS 113:277353–60
    [Google Scholar]
  4. Athey S, Imbens GW. 2019. Machine learning methods that economists should know about. Annu. Rev. Econ. 11:685–725
    [Google Scholar]
  5. Athey S, Tibshirani J, Wager S. 2019. Generalized random forests. Ann. Stat. 47:21148–78
    [Google Scholar]
  6. Athey S, Wager S. 2019. Estimating treatment effects with causal forests: an application. arXiv:1902.07409 [stat.ME]
  7. Auffhammer M, Rubin E 2018. Natural gas price elasticities and optimal cost recovery under consumer heterogeneity: evidence from 300 million natural gas bills NBER Work. Pap. 24295
  8. Baylis P. 2020. Temperature and temperament: evidence from twitter. J. Public Econ. 184:104161
    [Google Scholar]
  9. Baylis P, Obradovich N, Kryvasheyeu Y, Chen H, Coviello L et al. 2018. Weather impacts expressed sentiment. PLOS ONE 13:4e0195750
    [Google Scholar]
  10. Belloni A, Chen D, Chernozhukov V, Hansen C. 2012. Sparse models and methods for optimal instruments with an application to eminent domain. Econometrica 80:62369–429
    [Google Scholar]
  11. Belloni A, Chernozhukov V. 2011. -penalized quantile regression in high-dimensional sparse models. Ann. Stat. 39:182–130
    [Google Scholar]
  12. Belloni A, Chernozhukov V. 2013. Least squares after model selection in high-dimensional sparse models. Bernoulli 19:2521–47
    [Google Scholar]
  13. Belloni A, Chernozhukov V, Fernandez-Val I, Hansen C. 2017. Program evaluation and causal inference with high-dimensional data. Econometrica 85:1233–98
    [Google Scholar]
  14. Belloni A, Chernozhukov V, Hansen C. 2014a. High-dimensional methods and inference on structural and treatment effects. J. Econ. Perspect. 28:229–50
    [Google Scholar]
  15. Belloni A, Chernozhukov V, Hansen C. 2014b. Inference on treatment effects after selection among high-dimensional controls. Rev. Econ. Stud. 81:2608–50
    [Google Scholar]
  16. Bollinger B, Burkhardt J, Gillingham KT. 2020. Peer effects in residential water conservation: evidence from migration. Am. Econ. J. Econ. Policy 12:3107–33
    [Google Scholar]
  17. Boomhower J, Davis L. 2020. Do energy efficiency investments deliver at the right time?. Am. Econ. J. Appl. Econ. 12:1115–39
    [Google Scholar]
  18. Bound J, Jaeger DA, Baker RM. 1995. Problems with instrumental variables estimation when the correlation between the instruments and the endogeneous explanatory variable is weak. J. Am. Stat. Assoc. 90:430443–50
    [Google Scholar]
  19. Breiman L. 1996. Bagging predictors. Mach. Learn. 24:123–40
    [Google Scholar]
  20. Breiman L. 2001. Random forest. Mach. Learn. 45:5–32
    [Google Scholar]
  21. Breiman L, Friedman J, Stone C, Olshen R. 1984. Classification and Regression Trees Boca Raton, FL: Chapman and Hall/CRC
  22. Bühlmann P, van de Geer S. 2010. Statistics for High-Dimensional Data: Methods, Theory and Applications New York: Springer
  23. Bühlmann P, Yu B 2003. Boosting with the l2 loss. J. Am. Stat. Assoc. 98:462324–39
    [Google Scholar]
  24. Burkhardt J, Gillingham K, Kopalle PK. 2019. Experimental evidence on the effect of information and pricing on residential electricity consumption NBER Work. Pap. 25576
  25. Burlig F, Knittel C, Rapson D, Reguant M, Wolfram C 2020. Machine learning from schools about energy efficiency. J. Assoc. Environ. Resour. Econ. 7:61181–217
    [Google Scholar]
  26. Chen X, White H. 1999. Improved rates and asymptotic normality for nonparametric neural network estimators. IEEE Trans. Inform. Theory 45:2682–91
    [Google Scholar]
  27. Cheng B, Titterington DM. 1994. Neural networks: a review from a statistical perspective. Stat. Sci. 9:12–30
    [Google Scholar]
  28. Chernozhukov V, Chetverikov D, Demirer M, Duflo E, Hansen C et al. 2018. Double/debiased machine learning for treatment and structural parameters. Econom. J. 21:1C1–68
    [Google Scholar]
  29. Chernozhukov V, Chetverikov D, Demirer M, Duflo E, Hansen C, Newey W. 2017. Double/debiased/Neyman machine learning of treatment effects. Am. Econ. Rev. 107:5261–65
    [Google Scholar]
  30. Chernozhukov V, Newey W, Singh R. 2020. De-biased machine learning of global and local parameters using regularized Riesz representers. arXiv:1802.08667 [stat.ML]
  31. Chin S, Kahn ME, Moon HR. 2020. Estimating the gains from new rail transit investment: a machine learning tree approach. Real Estate Econ 48:3886–914
    [Google Scholar]
  32. Cunningham B, LaRiviere J, Wichman CJ. 2020. Clustered into control: heterogeneous causal impacts of water infrastructure failure. Econ. Inq https://doi.org/10.1111/ecin.12975
    [Crossref] [Google Scholar]
  33. Davuluri S, Francheschini RG, Knittel CR, Onda C, Roache K. 2019. Machine learning for solar accessibility: implications for low-income solar expansion and profitability NBER Work. Pap. 26178
  34. De Giorgi G, Harding M, Vasconcelos G. 2020. Predicting mortality from credit reports. Tech. Rep., Univ. Calif. Irvine, CA:
  35. Farrell MH. 2015. Robust inference on average treatment effects with possibly more covariates than observations. J. Econom. 189:11–23
    [Google Scholar]
  36. Farrell MH, Liang T, Misra S. 2018. Deep neural networks for estimation and inference. arXiv:1809.09953 [econ.EM]
  37. Faruqui A, Sergici S, Sharif A 2010. The impact of informational feedback on energy consumption: a survey of the experimental evidence. Energy 35:41598–608
    [Google Scholar]
  38. Fowlie M, Rubin E, Walker R 2019. Bringing satellite-based air quality estimates down to earth. AEA Pap. Proc. 109:283–88
    [Google Scholar]
  39. Friedman JH. 2001. Greedy function approximation: a gradient boosting machine. Ann. Stat. 29:51189–232
    [Google Scholar]
  40. Friedman JH, Hastie T, Tibshirani R. 2000. Additive logistic regression: a statistical view of boosting. Ann. Stat. 28:2337–407
    [Google Scholar]
  41. Friedman JH, Hastie T, Tibshirani R. 2010. Regularization paths for generalized linear models via coordinate descent. J. Stat. Softw. 33:1 https://doi.org/10.18637/jss.v033.i01
    [Crossref] [Google Scholar]
  42. Gentzkow M, Kelly B, Taddy M. 2019. Text as data. J. Econ. Lit. 57:3535–74
    [Google Scholar]
  43. Ghanem D, Shen S, Zhang J. 2020. A censored maximum likelihood approach to quantifying manipulation in China's air pollution data. J. Assoc. Environ. Resour. Econ. 7:5965–1003
    [Google Scholar]
  44. Ghoddusi H, Creamer GG, Rafizadeh N. 2019. Machine learning in energy economics and finance: a review. Energy Econ 81:C709–27
    [Google Scholar]
  45. Guha R, Ng S. 2019. A machine learning analysis of seasonal and cyclical sales in weekly scanner data NBER Work. Pap. 25899
  46. Harding M, Hersh J 2018. Big data in economics. IZA World Labor 2018:451
    [Google Scholar]
  47. Harding M, Hsiaw A. 2014. Goal setting and energy conservation. J. Econ. Behav. Organ. 107:209–27
    [Google Scholar]
  48. Harding M, Kettler K, Lamarche C, Ma L. 2021. Environmental and social benefits of a randomized dynamic pricing experiment. Work. Pap., Univ. Calif. Irvine, CA:
  49. Harding M, Lamarche C. 2016. Empowering consumers through data and smart technology: experimental evidence on the consequences of time-of-use electricity pricing policies. J. Policy Anal. Manag. 35:4906–31
    [Google Scholar]
  50. Harding M, Lamarche C. 2019. A panel quantile approach to attrition bias in big data: evidence from a randomized experiment. J. Econom. 211:161–82
    [Google Scholar]
  51. Harding M, Lamarche C, Pesaran MH. 2020. Common correlated effects estimation of heterogeneous dynamic panel quantile regression models. J. Appl. Econom. 35:3294–314
    [Google Scholar]
  52. Harding M, Sexton S. 2017. Household response to time-varying electricity prices. Annu. Rev. Resour. Econ. 9:337–59
    [Google Scholar]
  53. Hastie T, Tibshirani R, Friedman J. 2009. The Elements of Statistical Learning: Data Mining, Inference and Prediction. New York: Springer
  54. Heft-Neal S, Burney J, Bendavid E, Voss KK, Burke M. 2020. Dust pollution from the Sahara and African infant mortality. Nat. Sustain. 3:863–71
    [Google Scholar]
  55. Hoerl AE, Kennard RW. 1970. Ridge regression: biased estimation for nonorthogonal problems. Technometrics 12:155–67
    [Google Scholar]
  56. Hornik K, Stinchcombe M, White H. 1989. Multilayer feedforward networks are universal approximators. Neural Netw 2:5359–66
    [Google Scholar]
  57. Imbens GW, Rubin DB. 2015. Causal Inference for Statistics, Social, and Biomedical Sciences Cambridge, UK: Cambridge Univ. Press
  58. Ito K. 2014. Do consumers respond to marginal or average price? Evidence from nonlinear electricity pricing. Am. Econ. Rev. 104:2537–63
    [Google Scholar]
  59. Jessoe K, Rapson D. 2014. Knowledge is (less) power: experimental evidence from residential energy use. Am. Econ. Rev. 4:1041417–38
    [Google Scholar]
  60. Kock AB. 2013. Oracle efficient variable selection in random and fixed effects panel data models. Econom. Theory 29:1115–52
    [Google Scholar]
  61. Kock AB. 2016. Oracle inequalities, variable selection and uniform inference in high-dimensional correlated random effects panel data models. J. Econom. 195:171–85
    [Google Scholar]
  62. Koenker R. 2004. Quantile regression for longitudinal data. J. Multivar. Anal. 91:74–89
    [Google Scholar]
  63. Kuan CM, White H. 1994. Artificial neural networks: an econometric perspective. Econom. Rev. 13:11–91
    [Google Scholar]
  64. La Nauze A 2019. Power from the people: rooftop solar and a downward-sloping supply of electricity. J. Assoc. Environ. Resour. Econ. 6:61135–68
    [Google Scholar]
  65. Lamarche C. 2010. Robust penalized quantile regression estimation for panel data. J. Econom. 157:2396–408
    [Google Scholar]
  66. LeCun Y, Bengio Y, Hinton G. 2015. Deep learning. Nature 521:436–44
    [Google Scholar]
  67. Leeb H, Pötscher BM. 2006. Can one estimate the conditional distribution of post-model-selection estimators?. Ann. Stat. 34:52554–91
    [Google Scholar]
  68. Luo Y, Spindler M. 2017a. Estimation and inference of treatment effects with L2-boosting in high-dimensional settings. arXiv:1801.00364 [stat.ML]
  69. Luo Y, Spindler M. 2017b. L2-boosting for economic applications. Am. Econ. Rev. 107:5270–73
    [Google Scholar]
  70. Mullainathan S, Spiess J. 2017. Machine learning: an applied econometric approach. J. Econ. Perspect. 31:287–106
    [Google Scholar]
  71. Novan K, Smith A, Zhou T. 2020. Residential building codes do save energy: evidence from hourly smart-meter data. Rev. Econ. Stat https://doi.org/10.1162/rest_a_00967
    [Crossref] [Google Scholar]
  72. Prest BC. 2020. Peaking interest: how awareness drives the effectiveness of time-of-use electricity pricing. J. Assoc. Environ. Resour. Econ. 7:1103–43
    [Google Scholar]
  73. Segal MR. 1992. Tree-structured methods for longitudinal data. J. Am. Stat. Assoc. 87:418407–18
    [Google Scholar]
  74. Sela R, Simonoff J. 2012. RE-EM trees: a data mining approach for longitudinal and clustered data. Mach. Learn. 86:169–207
    [Google Scholar]
  75. Semenova V, Goldman M, Chernozhukov V, Taddy M. 2021. Estimation and inference on heterogeneous treatment effects in high-dimensional dynamic panels. arXiv:1712.09988 [stat.ML]
  76. Staiger D, Stock JH. 1997. Instrumental variables regression with weak instruments. Econometrica 65:3557–86
    [Google Scholar]
  77. Su L, Ju G. 2018. Identifying latent grouped patterns in panel data models with interactive fixed effects. J. Econom. 206:2554–73
    [Google Scholar]
  78. Su L, Shi Z, Phillips PCB. 2016. Identifying latent structures in panel data. Econometrica 84:62215–64
    [Google Scholar]
  79. Tibshirani R. 1996. Regression shrinkage and selection via the lasso. J. R. Stat. Soc. Ser. B 58:1267–88
    [Google Scholar]
  80. Wager S, Athey S. 2018. Estimation and inference of heterogeneous treatment effects using random forests. J. Am. Stat. Assoc. 113:5231228–42
    [Google Scholar]
  81. White H. 1992. Artificial Neural Networks: Approximation and Learning Theory Hoboken, NJ: Blackwell
  82. Zou H, Hastie T. 2005. Regularization and variable selection via the elastic net. J. R. Stat. Soc. Ser. B 67:2301–20
    [Google Scholar]
/content/journals/10.1146/annurev-resource-100920-034117
Loading
/content/journals/10.1146/annurev-resource-100920-034117
Loading

Data & Media loading...

Supplemental Material

Supplementary Data

  • Article Type: Review Article
This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error