Comparative Study of Mathematical Models for Ghana’s Gold Production

Abstract

Sigmoid functions were used to approximate the cumulative gold production in Ghana. The functions examined were the logistic, Gompertz, Gaussian, Probit and the Hill, which were then used to predict into the future. Although all the five approximated models gave a good estimation of the reality, the Gompertz function was identified to give the best approximation of the observed trend of gold production in Ghana. The model was selected based on its high proportion of variance explained (R2 = 0.9402) as well as having the least value in terms of error (RMSE) and information loss (AIC). The model suggested that gold production could escalate to a level of 7040813 Oz in the future. It also showed that, industrial scale production of gold should be possible even after the year 2050 provided the current upward trend of gold production in Ghana continues. Keywords: Sigmoid, Logistic, Gompertz, Gaussian, Probit, Hill, Gold