EconBase
← All papers

A Unifying Framework for Robust and Efficient Inference with Unstructured Data

Jacob Carlson, Melissa Dell

arXiv 1 May 2025 · Econometrics

arXiv:2505.00282 · PDF · Extracted main text

Abstract

This paper presents a general framework for conducting efficient inference on parameters derived from unstructured data, which include text, images, audio, and video. Economists have long used unstructured data by first extracting low-dimensional structured features (e.g., the topic or sentiment of a text), since the raw data are too high-dimensional and uninterpretable to include directly in empirical analyses. The rise of deep neural networks has accelerated this practice by greatly reducing the costs of extracting structured data at scale, but neural networks do not make generically unbiased predictions. This potentially propagates bias to the downstream estimators that incorporate imputed structured data, and the availability of different off-the-shelf neural networks with different biases moreover raises p-hacking concerns. To address these challenges, we reframe inference with unstructured data as a problem of missing structured data, where structured variables are imputed from high-dimensional unstructured inputs. This perspective allows us to apply classic results from semiparametric inference, leading to estimators that are valid, efficient, and robust. We formalize this approach with MAR-S, a framework that unifies and extends existing methods for debiased inference using machine learning predictions, connecting them to familiar problems such as causal inference. Within this framework, we develop robust and efficient estimators for both descriptive and causal estimands and address challenges like inference with aggregated and transformed missing structured data-a common scenario that is not covered by existing work. These methods-and the accompanying implementation package-provide economists with accessible tools for constructing unbiased estimators using unstructured data in a wide range of applications, as we demonstrate by re-analyzing several influential studies.

Citation extraction

109
references
266
in-text mentions
109
distinct cited
4
self-citations
21,067
main-text words

appendix boundary found by appendix_titled_section at “Supplemental Results” · 68% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Chernozhukov, Victor and Chetverikov, Denis and Demirer, Mert and Du… (2018) Double/debiased machine learning for treatment and structural parameters1.00074100%
2Robins, James M. and Rotnitzky, Andrea and Zhao, Lue Ping (1994) Estimation of Regression Coefficients When Some Regressors Are Not Always Observed1.00063100%
3Rubin, Donald B (1976) Inference and Missing Data1.00063100%
4Chernozhukov, Victor and Newey, Whitney K. and Singh, Rahul (2022) Automatic Debiased Machine Learning of Causal and Structural Effects1.00053100%
5Chernozhukov, Victor and Escanciano, Juan Carlos and Ichimura, Hideh… (2022) Locally Robust Semiparametric Estimation1.00053100%
6Zrnic, Tijana and Candès, Emmanuel J (2024) Active Statistical Inference0.9507486%
7Kennedy, Edward H (2023) Semiparametric doubly robust targeted double machine learning: a review0.9507386%
8Chen, Xiaohong and Hong, Han and Tarozzi, Alessandro (2008) Semiparametric efficiency in GMM models with auxiliary data0.9285480%
9Chernozhukov, Victor and Newey, Whitney and Quintas-Martinez, Víctor… (2022) RieszNet and ForestRiesz: Automatic Debiased Machine Learning with Neural Nets and Random Forests0.92843100%
10Imbens, Guido W. and Rubin, Donald B (2015) Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction0.92843100%

Showing the top 10 of 109 scored citations.

Cited by, within the corpus

arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.

Citing paperIntensityMentionsSections
1Large Language Models: An Applied Econometric Framework0.64422
2Making Interpretable Discoveries from Unstructured Data: A High-Dimensional Multiple Hypothesis Testing Approach0.64422
3Econometrics with Pre-Trained Embeddings for Unstructured Data0.51121
4Program Evaluation with Remotely Sensed Outcomes0.40511
5On Local Overidentification and Efficiency Gains in Modern Causal Inference and Data Combination0.40511
6From Unstructured Data to Demand Counterfactuals: Theory and Practice0.40511
7Bootstrapping with AI/ML-generated labels0.40511