GammaModel: GammaModel

Description Details References

Description

Tools to model count data in discrete classes using the Gamma distribution

Details

Package to analyse age-at-death count data across discrete age classes.

Inferential tools where there are large uncertainties from a) small sample sizes, and b) multi-class assignments. Therefore typically applicable to Archaeological data.

Calculates the exact probability of observed data given either a Gamma model or an age-class model. Therefore unsuitable for data with large sample sizes in multi- class assignments, since all possible combinations are computed.

MLE (under either model) found using DEoptimR search algorithm.

Likelihood distribution (under Gamma model) found using MCMC Metropolis-Hastings algorithm.

Performs Goodness of Fit tests, and model comparison using AIC and BIC.

References

'Modelling caprine age-at-death profiles using the Gamma distribution' by Adrian Timpson, Rosalind E. Gillis, Katie Manning and Mark G. Thomas, published in Journal of Archaeological Science 2018


UCL/GammaModel documentation built on June 25, 2020, 5:38 a.m.