A problem-solving approach to data analysis for economics

Abstract

Data analysis for formal methods is constrained due to the lengthy dominance of the econometric view within economics. Best practice in statistics suggests a shift in emphasis from making statements about the sampling distribution of numerical data summaries to seeking data summaries that communicate well. The process philosophy perspective informing the original institutionalists and also evident in the tradition of Keynes is amenable to drawing from current developments in the field of statistics toward this goal. Compared to the econometric approach, it emphasizes data analysis over statistical inference, problem- solving over theory testing, and algorithmic over analytic mathematics. In the choice of tools made possible by current technology, it favors general purpose tools that are adaptable. It favors the instrumental efficacy of computational thinking, visualization, exploration, and discovery over the ceremonial aspects of the mathematical rhetoric of economics. It also encourages the attention to ethics and assumptions stressed by statisticians. Our aim is to provide an overview of the philosophical foundation and intellectual history of an alternative to the econometric view and to give some examples of how it might be applied to the data needs of formal methods for social economics.

Keywords:

• Process philosophy
• computational thinking
• pragmatism
• methodology
• data analysis

AMS Subject Classifications:

• B40
• C18

Supplementary data

The underlying research materials for this article can be accessed at http://statlive.org/fse15.

Notes

No potential conflict of interest was reported by the authors.

This article is part of the Special Issue on Formal Methods for Integrated Socio-Economic Analysis edited by Tara Natarajan. We thank the editor and two anonymous referees for invaluable comments.

1 See for example the American Statistical Association’s Ethical Guidelines for Statistical Practice (Committee on Professional Ethics , 1999). Vardeman and Morris (2003) go even further in their advice to young statisticians, echoing the admonition of Charles Sanders Peirce that science is a mode of life ethically devoted to finding the truth (Thellefsen & Sorensen , 2014, p. 439).

2 Of this, Hicks (1976) writes:

It runs in terms of demand curves, and supply curves and cost curves—just the old tools of equilibrium economics. ... ; but what a muddle he made for his successors!... The equilibrists, therefore, did not know that they were beaten; or rather ... they did not know that they had been challenged. (Hicks , 1976, p. 141)

3 In a very concrete example showing the relevance of John Dewey’s philosophy of cognition Hickman (2013, p. 66) offers this quote from Mark Johnson

if babies are learning the meaning of things and events, and if babies are not yet formulating propositions, then meaning and understanding must involve a great deal more than the ability to create and understand propositions and their corresponding linguistic utterances. ... Meaning traffics in patterns, images, qualities, feelings, and eventually concepts and propositions. (Johnson , 2007, pp. 8–9)

4 Colander, Holt, and Rosser (2004) devote an article to describing the concept of “cutting edge economics.” The group of statisticians cited here help delineate something analogous in statistics, though that discipline is far more pluralistic in its methodological filtering.