Improved parameter estimation and uncertainty propagation in Bayesian Radiocarbon-dated Event Count [REC] models

William Christopher Carleton and Dave Campbell
Additional contact information
William Christopher Carleton: Max Planck Institute for Chemical Ecology

No 7p9vx, OSF Preprints from Center for Open Science

Abstract: Data about the past contain chronological uncertainty that needs to be accounted for in statistical models. Recently a method called Radiocarbon-dated Event Count (REC) modelling has been explored as a way to improve the handling of chronological uncertainty in the context of statistical regression. REC modelling has so far employed a Bayesian hierarchical framework for parameter estimation to account for chronological uncertainty in count series of radiocarbon-dates. This approach, however, suffers from a couple of limitations. It is computationally inefficient, which limits the amount of chronological uncertainty that can be accounted for, and the hierarchical framework can produce biased, but highly precise parameter estimates. Here we report the results of an investigation in which we compared hierarchical REC models to an alternative with simulated data and a new R package called "chronup". Our results indicate that the hierarchical framework can produce correct high-precision estimates given enough data, but it is susceptible to sampling bias and has an inflated Type I error rate. In contrast, the alternative better handles small samples and fully propagates uncertainty into parameter estimates. In light of these results, we think the alternative method is more generally suitable for Palaeo Science applications.

Date: 2021-12-16
References: Add references at CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
https://osf.io/download/61bb114f3ebfa506d6b11da3/

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:osf:osfxxx:7p9vx

DOI: 10.31219/osf.io/7p9vx

Access Statistics for this paper

More papers in OSF Preprints from Center for Open Science
Bibliographic data for series maintained by OSF ().