Combine function

[Andreu Monforte-Barberan]

Hi everyone,

I am trying to understand the mathematical workings of the Combine function, and I don’t understand how the function r(t)=p_1​(t)·p_2​(t) is interpreted. I would say that to calculate the probability that the dates are simultaneous, we would need to integrate this function over the entire domain. That said, this doesn't seem to be what the text is suggesting. Moreover, doing that assumes that simultaneity is only within the same year, which seems too stong for such distant events. Could someone explain to me exactly how Combine works? What is the purpose of r(t)? How is the p-value calculated from it? 

Thank you very much.

Andreu Monforte-Barberán

[Christopher Ramsey]

Dear Andreu

It is just a renormalised product of the two probability density functions. If you are dealing with Normal distributions this is identical to the error weighted mean, and the standard error of the weighted mean. The assumption is that both estimates are for the same event - that is there is only one age you are trying to determine.

Best wishes

Christopher

> --
> 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 on the web visit https://groups.google.com/d/msgid/oxcal/0909e776-3414-4e2e-b2ba-ea286884d798n%40googlegroups.com.

[Bayliss, Alex]

In practice, there are only a few archaeological situations where you can be pretty certain that two 14C measurements that are not replicate results on the same organism are of the same actual date.

Examples are:

'Perfect' pairs of samples from the same grave (e.g. an articulated skeleton and an articulated set of animal bones placed as an offering in the grave).

The last firing of a 'corn-dryer' or 'malting kiln' where all the grain in a deposit was part of the same processing event.

Hope this helps,

Alex

Work with us to champion heritage and improve lives. Read our Future Strategy and get involved at ​​​​historicengland.org.uk/strategy. Follow us:  Facebook  |  Twitter  |  Instagram     Sign up to our newsletter      This e-mail (and any attachments) is confidential and may contain personal views which are not the views of Historic England unless specifically stated. If you have received it in error, please delete it from your system and notify the sender immediately. Do not use, copy or disclose the information in any way nor act in reliance on it. Any information sent to Historic England may become publicly available. We respect your privacy and the use of your information. Please read our full privacy policy for more information.

-----Original Message-----
From: 'Christopher Ramsey' via OxCal <ox...@googlegroups.com>
Sent: 17 September 2024 15:24
To: OxCal group <ox...@googlegroups.com>
Subject: Re: Combine function

-- WARNING: This is an external message. Please use caution when replying, opening attachments or clicking on any links in this e-mail.--

To view this discussion on the web visit https://groups.google.com/d/msgid/oxcal/ED088F00-6EA5-467F-9680-EE9E49643FBD%40arch.ox.ac.uk.

[Christopher Ramsey]

Yes - and even for replicates you should use R_Combine so that the calibration is done on the combination. Combine can be used with other dating techniques or combinations of techniques. In general if we have a parameter with two independent likelihood functions this is how they would both be used to update the prior - it is not specifically a chronological method.

Best wishes

Christopher

> On 17 Sep 2024, at 15:28, Bayliss, Alex <Alex.B...@HistoricEngland.org.uk> wrote:
>
> In practice, there are only a few archaeological situations where you can be pretty certain that two 14C measurements that are not replicate results on the same organism are of the same actual date.
>
> Examples are:
>
> 'Perfect' pairs of samples from the same grave (e.g. an articulated skeleton and an articulated set of animal bones placed as an offering in the grave).
>
> The last firing of a 'corn-dryer' or 'malting kiln' where all the grain in a deposit was part of the same processing event.
>
> Hope this helps,
>
> Alex
>
>
>

> <image537309.jpg>Work with us to champion heritage and improve lives. Read our Future Strategy and get involved at ​​​​historicengland.org.uk/strategy.

> To view this discussion on the web visit https://groups.google.com/d/msgid/oxcal/LO4P265MB38396CC1E578C68EAE9B7B39A5612%40LO4P265MB3839.GBRP265.PROD.OUTLOOK.COM.

[Andreu Monforte-Barberan]

  Thank you so much for the feedback. I truly appreciate it.