Bayesian Demographic Reconstruction in the US Southwest: “Playing” with Priors

Summary

This is an abstract from the "SAA 2024: Individual Abstracts" session, at the 89th annual meeting of the Society for American Archaeology.

Paleodemographic reconstruction is an essential prerequisite for understanding human ecology of ancient societies. In the US Southwest several studies have employed Bayesian statistical methods to improve population estimates. This paper compares two alternative implementations of Bayesian statistics to demographic reconstruction in the US Southwest – Empirical Bayesian Analysis and a more generic Uniform Probability Density Analysis (UPDA). The two methods differ significantly in their approach to the construction of “Priors”, the creation of which is a point of contention for critics of Bayesian statistics. This study employs the UPDA approach for the first time in the VEP2N study area where previous work employed the Empirical methodology to assess the impact of this difference on final results, and to explore how alternative Prior construction within the UPDA approach affects final outcomes. The work concludes that Prior construction must be performed thoughtfully, and while the more generic UPDA approach does not produce results drastically different from the highly specific and customized Empirical approach, that is critically contingent on the specific selection of sites and ware types used for Prior construction. For example, alternative data selection for UDPA priors can generate three, rather than the two, demographic cycles of the Empirical method.

This Resource is Part of the Following Collections

• SAA 2024: Individual Abstracts   •
• Society for American Archaeology 89th Annual Meeting, New Orleans, LA (2024)

Keywords

Spatial Coverage

Record Identifiers

Abstract Id(s): 39379.0