Ultimately key to any kind of system. To what degree is the data valid and accurate? If we hope to learn and use decisions based upon the data. See below a reference to an HBS Working Knowledge Paper, with details at the link. Technical.
A General Theory of Identification by Iavor Bojinov and Guillaume Basse
Statistical inference teaches us how to learn from data, whereas identification analysis explains what we can learn from it. This paper proposes a simple unifying theory of identification, encouraging practitioners to spend more time thinking about what they can estimate from the data and assumptions before trying to estimate it.
Author Abstract
What does it mean to say that a quantity is identifiable from the data? Statisticians seem to agree on a definition in the context of parametric statistical models—roughly, a parameter θ in a model P = {Pθ : θ ∈ Θ} is identifiable if the mapping θ 7→ Pθ is injective. This definition raises important questions: Are parameters the only quantities that can be identified? Is the concept of identification meaningful outside of parametric statistics? Does it even require the notion of a statistical model? Partial and idiosyncratic answers to these questions have been discussed in econometrics, biological modeling, and in some subfields of statistics like causal inference. This paper proposes a unifying theory of identification that incorporates existing definitions for parametric and nonparametric models and formalizes the process of identification analysis. The applicability of this framework is illustrated through a series of examples and two extended case studies.
Paper Information
Full Working Paper Text (pdf)
Working Paper Publication Date: February 2020
HBS Working Paper Number: HBS Working Paper #20-086
Faculty Unit(s): Technology and Operations Management ... "
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment