Answers are grounded in this paper and cite their sources.
The linear regression model is the dominant tool employed in applied risk and insurance research. Based on my 2016 APRIA lecture at Chengdu, China, I illustrate the simple geometry of the linear regression model, as well as some standard results from it: omitted variable bias (OVB), classical measurement error (CME), simultaneous equation models (SEM), and instrumental variable estimation. Instrumental variable estimation solves OVB, CME, and SEM problems by constructing similar triangles to retrieve consistent estimates. I apply these tools by estimating the determinants of the Witt and Mehr awards given annually for Journal of Risk and Insurance articles, as two examples. The Witt vs. Mehr awards also contrasts short-term scholarly recognition (Witt) versus long-term scholarly recognition (Mehr). The comments made here apply to other paper awards, such as those presented by the Asian Pacific Journal of Risk and Insurance. I also present a simple index function based on the classical Gini index (hence, this new index is denoted as the regression gini index, RGI) useful for comparing two regression models, and apply this to explain the empirical difference between the determinants of the Witt and Mehr awards.
No takes yet. Share an insight, caveat, or question.
Richard Butler (2017) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: