Foundations of agnostic statistics / Peter M. Aronow and Benjamin T. Miller
Imprint
New York : Cambridge University Press, 2019
Descript
xviii, 298 pages : illustrations ; 23 cm
SUMMARY
"The last three decades have seen a marked change in the manner in which quantitative empirical inquiry in the social and health sciences is conducted. Sometimes dubbed the "credibility revolution," this change has been characterized by a growing acknowledgment that the evidence that researchers adduce for their claims is often predicated on unsustainable assumptions. Our understanding of statistical and econometric tools has needed to change accordingly. We have found that conventional textbooks, which often begin with incredible modeling assumptions, are not well suited as a starting point for credible research"-- Provided by publisher
CONTENT
Part I. Probability : Probalbility theory -- Summarizing distribution -- Part II. Statistics : Learning from random samples -- Regression -- Parametric models -- Part III. Identification : Missing data -- Causal inference