Premium
Statistical Problems in Ancient Numismatics
Author(s) -
Stam A.J.
Publication year - 1987
Publication title -
statistica neerlandica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 39
eISSN - 1467-9574
pISSN - 0039-0402
DOI - 10.1111/j.1467-9574.1987.tb01208.x
Subject(s) - numismatics , extant taxon , blank , sample (material) , estimator , mathematics , hammer , statistics , archaeology , history , engineering , mechanical engineering , chemistry , structural engineering , chromatography , evolutionary biology , biology
In antiquity coins were struck by placing a heated coin blank between two dies and hitting the upper die with a heavy blow of a hammer. When a die wore off, a new one was engraved. So dies intended to be the same show small differences that may be observed on extant coins as “the varieties”. Numismatists collect statistics of die varieties in as large a sample as possible of extant coins of a single issue they can find to study. From these statistics they want to estimate the number of dies used in producing the coin issue. To do this they need a parametric stochastic model describing how coins survive and become part of the sample. Several models are tried and tested against the data of two large samples, one of them new. (Few large samples exist). Estimators are reviewed and some related statistical problems studied.