glymour, clark and ramsey, joseph and wang, shuyan (2023) How PC Can Discover "Constitution". [Preprint]

Text How PC Discovers Consitution 1.26.24.docx Download (2MB)

Abstract

In an exchange with Gebharter, Casini and Baumgartner (B and C) conclude that the PC algorithm cannot reliably discover the “constitution” of materials or systems and has not been used to do so. We address their arguments, show sufficient conditions for PC to discover composition from data framed as Gebharter proposes, describe how in scientific practice using frameworks for data different from Gebharter’s PC has discovered constitution in two domains and two senses, and explain why PC is more reliable than standard methods used for causal or constitutional inference with non-experimental data. We address B and C’s examples in Appendix 2.

---

Export/Citation:

EndNote | BibTeX | Dublin Core | ASCII/Text Citation (Chicago) | HTML Citation | OpenURL

Social Networking:

Share |

---

Item Type:	Preprint
Creators:	Creators Email ORCID glymour, clark cg09@andrew.cmu.edu ramsey, joseph jdramsey@andrew.cmu.edu wang, shuyan shuyanw@andrew,cmu.edu
Keywords:	PC algorithm, Causality, Composition, Discovery
Subjects:	General Issues > Causation Specific Sciences > Probability/Statistics
Depositing User:	www xyz
Date Deposited:	19 Feb 2024 04:48
Last Modified:	19 Feb 2024 04:48
Item ID:	23101
Subjects:	General Issues > Causation Specific Sciences > Probability/Statistics
Date:	16 February 2023
URI:	https://philsci-archive.pitt.edu/id/eprint/23101

Monthly Views for the past 3 years

Monthly Downloads for the past 3 years

Plum Analytics

Actions (login required)

View Item