ABSTRACT: Measuring the height of a tree takes longer than measuring its diameter at breast height and often, only the heights of a subset of trees of known diameter are measured in the forest inventories. Since trees with the same diameter are not usually of the same height, even within the same stand. We proposed two hyperbolic height diameter models (HEHDM & HMMHDM) with a stochastic component added to the deterministic height diameter models. This approach mimics the natural variability of heights and therefore provides a more realistic height prediction, as demonstrated by the results of the Kolmogorov Smirnov test and Shapiro-Wilk test. The mean function of top height over the Dbh using the least squares parameter estimates predicted closely the observed values of top height in all the HD-models with better prediction from the proposed HD models and in general, all the HD-models considered predicted reasonable estimate over the entire range of DBH.