Comparing coin finds from Roman sites

What is it about?

This paper examines the claim by Reece (1995) that multivariate statistics fail to analyse assemblages of Roman coins from excavations successfully. The paper shows that cluster analysis and Correspondence Analysis do, in fact, work very well once one has identified the extreme outlier in the data, in this case the coins from Fishbourne Roman palace.

Featured Image

Why is it important?

This paper is important for showing the efficacy of multivariate analysis in coinage studies.