OpenBUGS and Bayesian tools for Archaeology
There were well over 50 published papers describing the application of Bayesian statistics to archaeology up to 2004 (see Mike Baxter’s Statistics in Archaeology for an very full list). The majority of the techniques described are not readily available to the archaeological community at large because of the problem of implementing them in suitable software. The exception is, of course, many of the methods applied to radiocarbon dating, which are implemented in BCal, OxCal and other software. This webpage aims to help fill the gap by supplying code for OpenBUGS (formerly WinBUGS) implementations of archaeological problems.
OpenBUGS is freely available software for the constructing Bayesian statistical models and evaluating them using Markov-Chain Monte-Carlo (MCMC) methods. Details are provided on the OpenBUGS Project web pages. OpenBUGS provides an ideal platform for implementing many of the Bayesian methods for archaeology. If you are an archaeologist who can write simple programs and understand the mathematics of the papers, and are prepared to learn enough about MCMC to check that the output is reliable, then this is for you. If you are a statistician looking for interesting problems then this is also for you.
To use the models supplied here you will need to download and install OpenBUGS, and read the manual and some of the references in it to comprehend MCMC methods. For a general introduction to MCMC I would recommend Gilks WR, Richardson S & Spiegelhalter DJ (Eds.) (1996) Markov chain Monte Carlo in Practice Chapman & Hall, London. For a very wide ranging discussion of Bayesian modelling with worked examples in WinBUGS see Congdon P (2001) Bayesian Statistical Modelling. Wiley, Chichester.
Most of the files are worked examples from the literature. I would be grateful for any comments or suggestions to improve the models given here. Contributions of new models are also welcome.
Download the files by right clicking and selecting your browser’s download or equivalent option. The titles link to odc file suitable for WinBUGS or OpenBUGS, and there are also text versions suitable for import into JAGS.
Buck, C.E., Cavanagh, W.G. and Litton, C.D. 1996: Bayesian Approach to Interpreting Archaeological Data. Chichester: Wiley.
Section 7.3 Simple disease incidence (text version)
Section 7.4 Consumerland revisited (text version)
Section 9.2 Simple radiocarbon calibration (text version)
Section 9.4 Radiocarbon calibration – Case Study I – St Veit-Klinglberg, Austria – stratification (text version)
Section 9.5 Radiocarbon calibration – Case Study II – Jama River Valley Ecuador (text version)
Section 9.6 Radiocarbon calibration – Case Study III – Stolford, England – wiggle matching (text version)
Section 9.7 Radiocarbon calibration – Case Study IV – Kastanas, Greece – wiggle matching (text version)
Section 9.8 Radiocarbon calibration – Case Study V – The Chancay culture of Peru (text version)
Buck, C.E., Litton, C.D. and Shennan, S.J. 1994: A case study in combining radiocarbon and archaeological information: the early Bronze-Age settlement of St. Veit-Klinglberg, Land Salzburg, Austria. Germania 2 427-447.
St Veit-Klinglberg (text version)
Christen, J.A. and Litton, C.D. 1995: A Bayesian approach to wiggle-matching. Journal of Archaeological Science 22 719-725.
Stolford Log – tree-ring wiggle matching (text version)
Kastanas, Greece – archaeological wiggle matching (text version)
Christen JA & Nicholls G (2000) Random walk radiocarbon calibration. Technical Report #457, Mathematics Department, University of Auckland, New Zealand.
available as a gzipped postscript file from Geoff Nicholls’ website.
Christen-Nicholls.odc (text version)
Zeidler, J.A., Buck, C.E. and Litton, C.D. 1998: The integration of archaeological phase information and radiocarbon results from the Jama River Valley, Ecuador: a Bayesian approach. Latin American Antiquity 9 160-179.
Jama River Valley (text version)
Survey data / Image analysis
Jump to my homepage.
This page created 25 September 2001. Last updated 16 May 2013.
© Andrew Millard 2013