Formal Theory in Demographic Temporal Frequency Analysis: Decomposing the TFD Data Generating Process

Summary

John Rick’s 1987 paper in American Antiquity presented the first systematic overview of theory underlying the "dates as data" approach (i.e., demographic temporal frequency analysis, dTFA), describing the general outline of a data generating process (DGP) linking paleopopulation dynamics to temporal distributions of archaeological materials (temporal frequency distributions, tfds). While research pursued in the dTFA framework has gained momentum over the intervening decades, questions regarding its reliability if not legitimacy have also emerged. Arguably, this critical pushback is the result of the tacit but persistent reluctance both of advocates and critics to give full formal expression to dTFA’s core theory. In turn, this reluctance has led to the inadvertent conflation of the program’s foundational principles and its special-case conditions, as well as to vague impressions regarding the degree to which non-demographic forces obscure demographic information in tfds. In response, I explicitly formalize Rick’s framework as a functional decomposition of the tfd DGP. Features of this formal theory are highlighted that can be productively leveraged in the service of demographic inference, specifically in the context of inverse uncertainty quantification (e.g., backward uncertainty propagation and sensitivity analysis). Both paleopopulation size and growth rate estimation are considered.

This Resource is Part of the Following Collections

• Society for American Archaeology 83rd Annual Meeting, Washington, DC (2018)

Keywords

Record Identifiers

Abstract Id(s): 20517