Do We Really Even Need Data?
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913221849382912 |
|---|---|
| author | Hoffman, Kentaro Salerno, Stephen Afiaz, Awan Leek, Jeffrey T. McCormick, Tyler H. |
| author_facet | Hoffman, Kentaro Salerno, Stephen Afiaz, Awan Leek, Jeffrey T. McCormick, Tyler H. |
| contents | As artificial intelligence and machine learning tools become more accessible, and scientists face new obstacles to data collection (e.g. rising costs, declining survey response rates), researchers increasingly use predictions from pre-trained algorithms as outcome variables. Though appealing for financial and logistical reasons, using standard tools for inference can misrepresent the association between independent variables and the outcome of interest when the true, unobserved outcome is replaced by a predicted value. In this paper, we characterize the statistical challenges inherent to this so-called ``inference with predicted data'' problem and elucidate three potential sources of error: (i) the relationship between predicted outcomes and their true, unobserved counterparts, (ii) robustness of the machine learning model to resampling or uncertainty about the training data, and (iii) appropriately propagating not just bias but also uncertainty from predictions into the ultimate inference procedure. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_08702 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Do We Really Even Need Data? Hoffman, Kentaro Salerno, Stephen Afiaz, Awan Leek, Jeffrey T. McCormick, Tyler H. Methodology Machine Learning As artificial intelligence and machine learning tools become more accessible, and scientists face new obstacles to data collection (e.g. rising costs, declining survey response rates), researchers increasingly use predictions from pre-trained algorithms as outcome variables. Though appealing for financial and logistical reasons, using standard tools for inference can misrepresent the association between independent variables and the outcome of interest when the true, unobserved outcome is replaced by a predicted value. In this paper, we characterize the statistical challenges inherent to this so-called ``inference with predicted data'' problem and elucidate three potential sources of error: (i) the relationship between predicted outcomes and their true, unobserved counterparts, (ii) robustness of the machine learning model to resampling or uncertainty about the training data, and (iii) appropriately propagating not just bias but also uncertainty from predictions into the ultimate inference procedure. |
| title | Do We Really Even Need Data? |
| topic | Methodology Machine Learning |
| url | https://arxiv.org/abs/2401.08702 |