Methodological Workflow

Atin Basuchoudhary, James Bang, John David () and Tinni Sen ()
Additional contact information
John David: Virginia Military Institute
Tinni Sen: Virginia Military Institute

Chapter Chapter 3 in Identifying the Complex Causes of Civil War, 2021, pp 29-47 from Springer

Abstract: Abstract This chapter describes a formal methodological sequence of events that allows us to interpret hitherto “black-box” machine learning algorithmic outcomes as the effect of some cause. While we apply this methodological workflow to understanding the causes of civil conflict, it can help answer many questions in economics and political science. We use the backdoor criteria to help determine causal impacts and introduce the idea of empirically informed covariate selection to guide the researcher to fulfill these criteria reasonably. Last we suggest that predictive accuracy is a critical element in highlighting the nature of the causal impact.

Keywords: Empirically Informed Covariate Selection (EICS); Recursive Feature Elimination (RFE); Risk variables; Confounder variables; Collider variables; Intermediate variables; Partial Dependence Plots (PDP); Bayesian Additive Regression Tree (BART) (search for similar items in EconPapers)
Date: 2021
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-030-81993-4_3

Ordering information: This item can be ordered from
http://www.springer.com/9783030819934

DOI: 10.1007/978-3-030-81993-4_3

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().