Is it possible to find a good point estimate of a calibrated radiocarbon date?

Abstract

The result from probabilistic calibration of a radiocarbon date is given in the form of a probability density function. Consequently, reporting a 68% or 95% confidence interval has became a commonly accepted practice. However, many users of 14C dates still try to present the results of calibration as a single point. This manner of presentation is often applied during the construction of age-depth models due to its convenience and simplicity. In this paper, the author tests whether it is possible to find a good point estimate of a calibrated 14C date. The idea of the tests is to compare, using computer simulation, the true value of the calendar age with the age calculated based on the probabilistic calibration of the 14C date and the method of finding the point estimate. The test is carried out for the following point estimates: mode, median, average, the central point of the confidence intervals, and the local mode inside the confidence intervals. The results show that none of these may be considered as a good estimate. © 2007 by the Arizona Board of Regents on behalf of the University of Arizona.