SZ. BURGER ALICE: LATE ROMAN MONEY CIRCULATION IN SOUTH-PANNONIA / Régészeti Füzetek II/22. (Magyar Nemzeti Múzeum Budapest, 1981
THE METHODS OF THE ANALYSIS
At this paragraph I give only the simplest results of the clustering, that is the very first clusters generated by the stepwise agglomerative process, and the points with extreme big distances. The comments follow later. I. MINTS VERSUS SITE S (see p. 14-16) Sites where the distribution of coins coming from the specific mints are the most similar: PECST + CEMET VESZP + OCSOD PECSC + GRVS1 (*) OCSEM + SAG VC (and here comes the first cluster of three members:) PECST + CEMET + SAGVT (the next point joins an earlier cluster of two members, too:) VESZP + OCSOD + MAROK PECST + CEMET + SAGVT + GRVS2 (**) (the next step joins the clusters (*) and (**) and so one...) The sites with the most remarkable dissimilarities are MERNY / ARP AS and ACS / practically from all the other sites. The dual question about the mints, the coins of which have the most similar distributions at sites: ROMA + AQU IL SISC + THESS , but following the clustering process you can see that these two clusters are very far from each other. II. EMPERORS VERSUS SITE S (see p. 17-19) Sites where the distributions of coins of the specific Emperors are the most similar: ARPAS + TEDEJ MAROK + OCSOD MAROK + OCSOD + KAZSO ARPAS + TEDEJ + OCSEM The most remarkable dissimilarity can be found between ACS and the other sites. The dual question gives the clusters: VANTI + VALNS VANTI + VALNS + GRATI CNUS2 + CPOLI CNUS2 + CPOLI + URBSR and so one. The CDINA + CGALL cluster is rather far from the remaining points. III. MINTS VERSUS EMPEROR S (see p. 20-21) Mints, the coins of which have the most similar distributions related to the Emperors: ROMA + AQUIL THESS + CPOLI and so one. The dual question gives the clusters: CTANS + CPOLI CGALL + GRATI CNUS2 + URBSR CIUS2 + IULIA CGALL + GRATI + VANTI CNUS2 + URBS2 + CNUS1 CNUS2 + URBSR + CNUS1 + CDINA joined now) CNUS2 + URBSR + CNUS1 + CDINA + CTANS + CPOLI and so one. .. 1 0
