Weighting dates within a phase

[Brett Parbus]

Hello,

First time posting in here, so please forgive any ignorance to the guidelines for posting.

Is there a way to weight individual dates in a phase so that one date has more overall power over the resulting boundary calculations? As an example, say that I have dates from various archaeological contexts, each of which contains a similar overall variety of ceramic styles, but in different quantities. Could I build a phase model where the association between dates in the phase is the presence of a distinct ceramic style, but weighted so that dates from contexts with a greater percentage of this ceramic style have more power over the results of the model?

In the attached example, say I wanted to build a phase for five dates associated with Ceramic Style 3. Could I program in a way to give Context 1 a weight of 3 and Context 2 a weight of 18, and so on?

[Christopher Ramsey]

Dear Brett

I don't think directly. In general multi-phase models are number independent because the number of dated samples should not influence the phase length.

You can introduce undated events within each phase, for each context - that might help make some contexts (with associated constraints) have more impact if you are cross-referencing the phase boundaries but in the end you are not giving the actual measurement from that phase more weight. There is a limit to this though - too many undated events is likely to make the model difficult to run.

Best wishes

Christopher

> <Hypothetical Phase.png>
>
>
> --
> You received this message because you are subscribed to the Google Groups "OxCal" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to oxcal+un...@googlegroups.com.
> To view this discussion visit https://groups.google.com/d/msgid/oxcal/ed0f03f8-9d42-4526-95ee-fd4cb0f3f2fen%40googlegroups.com.
> <Hypothetical Phase.png>

[MILLARD, ANDREW R.]

Dear Brett,

Perhaps you can explain a bit more what you are aiming to calculate. What are you hoping the (weighted) boundary will tell you about? Is it the start/end of the site, or of the ceramic style, or something else? Why do you think that dates from contexts with a higher proportion of Ceramic Style 3 are more important to what you are trying to determine?

Best wishes
Andrew
--
Prof. Andrew Millard
Department of Archaeology,
Durham University, UK
Email: A.R.M...@durham.ac.uk
Personal page: https://www.durham.ac.uk/staff/a-r-millard/
Dunbar 1650 MOOC: https://www.futurelearn.com/courses/battle-of-dunbar-1650
Durham research: http://www.durham.ac.uk/global-research-brochure


-----Original Message-----
From: 'Christopher Ramsey' via OxCal <ox...@googlegroups.com>
Sent: 06 November 2025 14:47
To: OxCal group <ox...@googlegroups.com>
Subject: Re: Weighting dates within a phase

[EXTERNAL EMAIL]

To view this discussion visit https://groups.google.com/d/msgid/oxcal/C691D4A7-E9EB-4460-B62F-042DE7E0772F%40arch.ox.ac.uk.