DoubleMLDeep: Estimation of Causal Effects with Multimodal Data
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911770086473728 |
|---|---|
| author | Klaassen, Sven Teichert-Kluge, Jan Bach, Philipp Chernozhukov, Victor Spindler, Martin Vijaykumar, Suhas |
| author_facet | Klaassen, Sven Teichert-Kluge, Jan Bach, Philipp Chernozhukov, Victor Spindler, Martin Vijaykumar, Suhas |
| contents | This paper explores the use of unstructured, multimodal data, namely text and images, in causal inference and treatment effect estimation. We propose a neural network architecture that is adapted to the double machine learning (DML) framework, specifically the partially linear model. An additional contribution of our paper is a new method to generate a semi-synthetic dataset which can be used to evaluate the performance of causal effect estimation in the presence of text and images as confounders. The proposed methods and architectures are evaluated on the semi-synthetic dataset and compared to standard approaches, highlighting the potential benefit of using text and images directly in causal studies. Our findings have implications for researchers and practitioners in economics, marketing, finance, medicine and data science in general who are interested in estimating causal quantities using non-traditional data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_01785 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | DoubleMLDeep: Estimation of Causal Effects with Multimodal Data Klaassen, Sven Teichert-Kluge, Jan Bach, Philipp Chernozhukov, Victor Spindler, Martin Vijaykumar, Suhas Machine Learning Artificial Intelligence Econometrics Methodology 62, 91 I.2.0 This paper explores the use of unstructured, multimodal data, namely text and images, in causal inference and treatment effect estimation. We propose a neural network architecture that is adapted to the double machine learning (DML) framework, specifically the partially linear model. An additional contribution of our paper is a new method to generate a semi-synthetic dataset which can be used to evaluate the performance of causal effect estimation in the presence of text and images as confounders. The proposed methods and architectures are evaluated on the semi-synthetic dataset and compared to standard approaches, highlighting the potential benefit of using text and images directly in causal studies. Our findings have implications for researchers and practitioners in economics, marketing, finance, medicine and data science in general who are interested in estimating causal quantities using non-traditional data. |
| title | DoubleMLDeep: Estimation of Causal Effects with Multimodal Data |
| topic | Machine Learning Artificial Intelligence Econometrics Methodology 62, 91 I.2.0 |
| url | https://arxiv.org/abs/2402.01785 |