Empirical Validation and Model Selection in Archaeological Simulation

Summary

Empirical validation is a key stage of any model development process and should provide an objective and quantitative account of the model performance. Yet, too often this stage plays a marginal role in the inferential exercise, with many discussions almost exclusively dedicated on the model building process. This paper discusses this neglected aspect of archaeological simulation, distinguishing two approaches drawn from epistemological parallels with statistical modelling. The first utilises simulations as some sort of null templates; observed summary statistics are compared against simulation generated data, and significant deviations often becomes the basis for the development of more complex models. The second approach, which is the main focus of this paper, is centred on the theoretical framework of multi-model inference. Multiple competing models are formalised as computer simulations and compared against each other on the basis of their complexity, knowledge of the parameters, and goodness-of-fit to the observed data. The result is a probabilistic evaluation of whether one model is "better" than another, a solution that can overcome the problem of equifinality through its quantification. Examples from Neolithic and Bronze Age Europe will showcase the pros and the cons of this latter approach.

SAA 2015 abstracts made available in tDAR courtesy of the Society for American Archaeology and Center for Digital Antiquity Collaborative Program to improve digital data in archaeology. If you are the author of this presentation you may upload your paper, poster, presentation, or associated data (up to 3 files/30MB) for free. Please visit http://www.tdar.org/SAA2015 for instructions and more information.

This Resource is Part of the Following Collections

• Simulating Social Complexity to Understand the Archaeological Past   •
• Society for American Archaeology 80th Annual Meeting, San Francisco, CA (2015)

Keywords

Spatial Coverage