In forensic research, stature is an important indicator in the identification of humans. There are numerous methods for estimating stature, and their goal is to determine the optimal variables for delivering the most accurate predictions. The purpose of this study is to compare the predictive algorithms for stature based on various hand dimensions. The selected hand variables can be separated into four categories-length, breadth, wrist, and thickness-and 18 variables were eventually selected in this research. The hand dimension data were analyzed by descriptive statistics. In the Korean population, there were significant differences found within genders in terms of hands and stature. Two predictive algorithms, regression and artificial neural network, were compared on the basis of their coefficient of determination (R2) and root mean square error (RMSE). In the single linear regression, hand length (R2 = .386) and palm length (R2 = .349) were found to be the most relevant variables in stature prediction for males. For females, hand length (R2 = .286) and inner grip circumference (R2 = .261) scored the highest R2. In the multiple linear regression, an R2 of .659 was obtained for both males and females, with an RMSE of 5.38 cm. In the artificial neural network, the value of R2 was .05, along with an RMSE of 5.17 cm. Overall, this study proposes the artificial neural networks as an improved predictive algorithm for stature, and hand length and inner grip circumference were found to be the most relevant variables to predict stature.