Direct and indirect effects--an information theoretic perspective
Information theoretic (IT) approaches to quantifying causal influences have experienced some popularity in the literature, in both theoretical and applied (e.g., neuroscience and climate science) domains. While these causal measures are desirable in that they are model agnostic and can capture non-l...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
Multidisciplinary Digital Publishing Institute
2020
|
| Online Access: | https://hdl.handle.net/1721.1/127685 |
| _version_ | 1826211966438342656 |
|---|---|
| author | Schamberg, Gabriel Chapman, William Xie, Shang-Ping Coleman, Todd P. |
| author2 | Picower Institute for Learning and Memory |
| author_facet | Picower Institute for Learning and Memory Schamberg, Gabriel Chapman, William Xie, Shang-Ping Coleman, Todd P. |
| author_sort | Schamberg, Gabriel |
| collection | MIT |
| description | Information theoretic (IT) approaches to quantifying causal influences have experienced some popularity in the literature, in both theoretical and applied (e.g., neuroscience and climate science) domains. While these causal measures are desirable in that they are model agnostic and can capture non-linear interactions, they are fundamentally different from common statistical notions of causal influence in that they (1) compare distributions over the effect rather than values of the effect and (2) are defined with respect to random variables representing a cause rather than specific values of a cause. We here present IT measures of direct, indirect, and total causal effects. The proposed measures are unlike existing IT techniques in that they enable measuring causal effects that are defined with respect to specific values of a cause while still offering the flexibility and general applicability of IT techniques. We provide an identifiability result and demonstrate application of the proposed measures in estimating the causal effect of the El Niño–Southern Oscillation on temperature anomalies in the North American Pacific Northwest. |
| first_indexed | 2024-09-23T15:14:14Z |
| format | Article |
| id | mit-1721.1/127685 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T15:14:14Z |
| publishDate | 2020 |
| publisher | Multidisciplinary Digital Publishing Institute |
| record_format | dspace |
| spelling | mit-1721.1/1276852022-10-02T01:33:36Z Direct and indirect effects--an information theoretic perspective Schamberg, Gabriel Chapman, William Xie, Shang-Ping Coleman, Todd P. Picower Institute for Learning and Memory Information theoretic (IT) approaches to quantifying causal influences have experienced some popularity in the literature, in both theoretical and applied (e.g., neuroscience and climate science) domains. While these causal measures are desirable in that they are model agnostic and can capture non-linear interactions, they are fundamentally different from common statistical notions of causal influence in that they (1) compare distributions over the effect rather than values of the effect and (2) are defined with respect to random variables representing a cause rather than specific values of a cause. We here present IT measures of direct, indirect, and total causal effects. The proposed measures are unlike existing IT techniques in that they enable measuring causal effects that are defined with respect to specific values of a cause while still offering the flexibility and general applicability of IT techniques. We provide an identifiability result and demonstrate application of the proposed measures in estimating the causal effect of the El Niño–Southern Oscillation on temperature anomalies in the North American Pacific Northwest. 2020-09-23T17:32:00Z 2020-09-23T17:32:00Z 2020-07 2020-08-21T13:50:57Z Article http://purl.org/eprint/type/JournalArticle 1099-4300 https://hdl.handle.net/1721.1/127685 Schamberg, Gabriel et al. "Direct and indirect effects--an information theoretic perspective." Entropy 22, 8 (July 2020): 854 ©2020 Author(s) 10.3390/e22080854 Entropy Creative Commons Attribution https://creativecommons.org/licenses/by/4.0/ application/pdf Multidisciplinary Digital Publishing Institute Multidisciplinary Digital Publishing Institute |
| spellingShingle | Schamberg, Gabriel Chapman, William Xie, Shang-Ping Coleman, Todd P. Direct and indirect effects--an information theoretic perspective |
| title | Direct and indirect effects--an information theoretic perspective |
| title_full | Direct and indirect effects--an information theoretic perspective |
| title_fullStr | Direct and indirect effects--an information theoretic perspective |
| title_full_unstemmed | Direct and indirect effects--an information theoretic perspective |
| title_short | Direct and indirect effects--an information theoretic perspective |
| title_sort | direct and indirect effects an information theoretic perspective |
| url | https://hdl.handle.net/1721.1/127685 |
| work_keys_str_mv | AT schamberggabriel directandindirecteffectsaninformationtheoreticperspective AT chapmanwilliam directandindirecteffectsaninformationtheoreticperspective AT xieshangping directandindirecteffectsaninformationtheoreticperspective AT colemantoddp directandindirecteffectsaninformationtheoreticperspective |