This paper presents the development of statistical 3D body shape prediction models and how these models can be shared to be used by other researchers, software developers or organisations. In Digital human modelling (DHM) tools it is important that the generated manikin models are accurate and representative for different body sizes and shapes. By using both 3D body scan data and one-dimensional data, provided in tabular format, a prediction model was created that based on a few input variables predicts both additional missing one-dimensional data and body shape data, in the form of landmark coordinates in XYZ point cloud format as well as a body shape mesh model in OBJ format. The data is handled in a sequential process where three different prediction functions were defined using similar implementations of a conditional regression model. The developed statistical 3D body shape prediction model described in this paper is based on body scan data from the CAESAR anthropometric survey. The statistical 3D body shape prediction models, consisting of MATLAB scripts have been prepared, packaged and shared in an online repository on GitHub under the MIT License. Since the model is shared under an open license, the idea and intention are that it can be further developed by other researchers and organizations. This first version only generates a static body shape whereas future versions could include joint center prediction models and deformation patterns in relation to joint angles, to enable accurate body deformation during different postures and motions.