2001
DOI: 10.1111/1467-9876.00250
|View full text |Cite
|
Sign up to set email alerts
|

Radiocarbon Dating with Temporal Order Constraints

Abstract: Bayesian methods are now widely used for analysing radiocarbon dates. We ®nd that the non-informative priors in use in the literature generate a bias towards wider date ranges which does not in general re¯ect substantial prior knowledge. We recommend using a prior in which the distribution of the difference between the earliest and latest dates has a uniform distribution. We show how such priors are derived from a simple physical model of the deposition and observation process. We illustrate this in a case-stu… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
61
0

Year Published

2004
2004
2014
2014

Publication Types

Select...
5
4

Relationship

2
7

Authors

Journals

citations
Cited by 51 publications
(61 citation statements)
references
References 17 publications
0
61
0
Order By: Relevance
“…In the case that there is a lower limit llim a for t a and an upper limit ulim m for t m the number of possible combinations of solutions is proportional to: .30) and so the reciprocal of this can be added as a prior factor (these latter two factors were suggested in Nicholls and Jones (2001) and Jones and Nicholls (2002) and are incorporated in OxCal prior to version 4 as described in Bronk Ramsey (2001)). If there is also an upper limit ulim a for t a and a lower limit llim m for t m the number of possible combinations becomes:…”
Section: A34 Multiple Segmentsmentioning
confidence: 99%
“…In the case that there is a lower limit llim a for t a and an upper limit ulim m for t m the number of possible combinations of solutions is proportional to: .30) and so the reciprocal of this can be added as a prior factor (these latter two factors were suggested in Nicholls and Jones (2001) and Jones and Nicholls (2002) and are incorporated in OxCal prior to version 4 as described in Bronk Ramsey (2001)). If there is also an upper limit ulim a for t a and a lower limit llim m for t m the number of possible combinations becomes:…”
Section: A34 Multiple Segmentsmentioning
confidence: 99%
“…In cases where there is a lot of dating information, the results depend only weakly on the assumptions embedded in the model of stratigraphy, but in cases where there are few dates or the intervals between events are comparable to the uncertainties in the dates, the effects of these assumptions need to be assessed (Steier and Rom, 2000;Nicholls and Jones, 2001). In addition, although the models capture the important information from stratigraphy of ordering events in time, this is done at the cost of introducing explicit statistical assumptions.…”
Section: Statistical Approachmentioning
confidence: 99%
“…In addition, although the models capture the important information from stratigraphy of ordering events in time, this is done at the cost of introducing explicit statistical assumptions. The uniform prior is a repeated and statistically simple assumption, but Nicholls and Jones (2001) showed that it implies a priori that long durations between events are much more probable than short ones. Usually, this has little effect on the results, as the chronometric data constrain the duration between events much more strongly than the prior stratigraphic model, but in situations with complex stratigraphic models and few chronometric data, it can affect the results.…”
Section: Statistical Approachmentioning
confidence: 99%
“…Nicholls and Jones (2001) provide a prior suitable for calibration using temporal order constraints with archaeological data, and we will use that prior here. (See Nicholls and Jones [2001] for a detailed discussion of model specifics.) Under their calibration scheme, all analyzed dates are treated as coming from one of a number of phases that occur as a single series, and temporal constraints on the timing of each phase are set on the basis of the analyzed data.…”
Section: ∆R = P Ma -Q Atmentioning
confidence: 99%