Sequential/overlapping phases in model and low agreement of individual determinations

[Daniel Sacramento Van Calker]

Hello everyone, hope you'll be able to help with the model below.

This model aims to characterize three different phases. They may overlap, but they are mostly sequential (in this example, I've only kept the first and last dates for each phase). What I'm currently struggling with is the low agreement of the two earliest dates in Phase 3. Archaeologically, they make sense within this phase, but they are earlier than the rest of the dataset and overlap with the end of Phase 2. I've run the model using the general outlier function for both dates, but their individual agreement indices remain quite low (<24). If I remove these dates, the results for the rest of the model become even worse. Any ideas on how to proceed? Thank you much for your time and attention.

Daniel van Calker

 Plot()
 {
  Curve("IntCal20","intcal20.14c");
  Outlier_Model("General",T(5),U(0,4),"t");
  Sequence("P")
  {
   Boundary("Start 1");
   Phase("1")
   {

    R_Date("Wk",4960,31);

....

    R_Date("Bt-7",4430,50);
   };
   Boundary("End 1");
   Sequence("U")
   {
    Boundary("Start 2");
    Phase("2")
    {
     R_Date(Bt-66,4490,30);

...

     R_Date("F5",4085,38);
    };
    Boundary("2");
    Sequence("PO")
    {
     Boundary("Start 3");
     Phase("3")
     {
      R_Date("Bt-24",4240,40)
      {
       Outlier("General",0.05);
      };
      R_Date("Bt-37",4240,30)
      {
       Outlier("General",0.05);
      };

...

      R_Date("Bt‑25",4000,40);
     };
     Boundary("End Pos Idolos Placa");
    };
   };
  };
 };

[Derek Hamilton]

Hi Daniel,

Bt-66 needs to be in quotes like the other dates… are they omitted in your actual model? Are others missing these? The syntax is very important.

Best wishes,

Derek

On Jun 23, 2026, at 5:52 PM, Daniel Sacramento Van Calker <daniel...@campus.ul.pt> wrote:

You don't often get email from daniel...@campus.ul.pt. Learn why this is important

--
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/cd9da36e-10b3-41f6-8148-23a17a59aab6n%40googlegroups.com.

[Erik Marsh]

Hi Daniel,

Derek's right, the missing quotes for Bt-66 is a problem; OxCal misreads this as 0±4490 years BP. Fixing this, your model does run but with low As. I would also check your organization of phases (easiest to see in List view). You have multiple nested sequences, which are usually not necessary. These sequences assume that all phase 3 dates were deposited after all phase 2 dates. If this depositional sequence is clear from stratigraphy, this may be correct, even with low As. However, if these are in fact overlapping phases, each should have its sequence, all grouped in an overall phase (as below; also, no need to call IntCal20 since this is the default). This runs fine with no low As.

Hope this helps

Erik

 

Phase()
{
 Sequence()

 {
  Boundary("Start 1");
  Phase("1")
  {
   R_Date("Wk",4960,31);

   R_Date("Bt-7",4430,50);
  };
  Boundary("End 1");

 };
 Sequence()

 {
  Boundary("Start 2");
  Phase("2")
  {
 R_Date("Bt-66",4490,30);

   R_Date("F5",4085,38);
  };
  Boundary("End 2");
 };
 Sequence()

 {
  Boundary("Start 3");
  Phase("3")
  {

   R_Date("Bt-24",4240,40);
   R_Date("Bt-37",4240,30);
   R_Date("Bt‑25",4000,40);
  };
  Boundary("End 3");
 };
};

[Daniel Sacramento Van Calker]

Hi Derek and Erik,

Thank you so much for your inputs! Yes they are in quotes, I've deleted them by accident when the code was copied. Also in the last boundary line it should be "  Boundary("End 3"); ", as Erik noticed. Yes, my assumption was that there was something wrong with the organization of phases. Archaeological data does suggest an overlap between phase 2 and 3 and that should be corrected and run fine with your suggestion!

Again, thank you for your valuable help on this.

Best wishes,

Daniel

To view this discussion visit https://groups.google.com/d/msgid/oxcal/ffd6aeb8-5022-420c-b0f7-9c48e374be23n%40googlegroups.com.

--

Daniel van Calker

PhD Student in Archaeology

UNIARQ - Centro de Arqueologia da Universidade de Lisboa

ResearchGate