Within this work, a generalized two-dimensional energetic hysteresis model for characterizing the magnetization process of thin ferromagnetic films is presented. Based on the -Energetic Model of Ferromagnetic Hysteresis- (EM), which has been introduced by H. Hauser in 1994, so called statistical domain classes adapted from the magnetic easy directions of the material sample are distinguished. With respect to Hauser-s formulation of the EM, two major extensions are provided in this work. As a first generalization, the orientation of the elementary magnetic dipoles within a statistical domain class is no longer restricted to the specified easy direction, but is represented by a stochastic circular distribution function that is characterized by a mean orientation and a certain variance. So local misalignments of magnetic moments due to imperfections within the material can be modeled on the one hand, and the temperature dependence of the spontaneous magnetization can be described from the first principles on the other hand. The second key extension is the fully two-dimensional formulation of the EM. Although, the original model of Hauser is intended for a certain number of easy directions, most of the investigations are based on a one-dimensional model setup in consideration of the symmetry properties of the crystalline axes. In particular, the terms for reversible and irreversible work during the magnetization process have to be reformulated in order to allow fully two-dimensional modeling. Further, the magnetocrystalline and induced anisotropy are directly incorporated in the model by corresponding energy terms. Thereby, the two-dimensional approach allows predicting the magnetic hysteresis curves of anisotropic thin films for all directions of the applied field with a single set of parameters. Besides, the generalized model accounts for coherent magnetization rotation as well as non-coherent magnetization reversal mechanisms by using a corresponding parameterization of domain classes. The evaluation of the presented generalized two-dimensional energetic hysteresis model is done by the example of Permalloy thin films used for anisotropic magneto-resistive (AMR) sensors. Especially for AMR sensor applications the directional properties of the magnetization curve are important, particularly the hard-axis characteristics. The model parameters are identified by a comparison of the simulated hysteresis loops to those obtained from a magneto-optical Kerr measurement setup in easy axis and hard axis direction. Then, the magnetization curves can be calculated for any arbitrary direction of the magnetic field with respect to the easy axis. Furthermore, the identified model parameters reflect several microstructural properties of the thin film, like microscopic misalignments due to inhomogeneities, for example. Amongst others, these microstructural properties are related to characteristics of the thin film production process, such as the distance between target and substrate during sputtering. Due to the fact that the magnetic hysteresis model presented within this work allows an inference from measured magnetization curves to the structural properties of the film, laborious and expensive microscopic analyses can be reduced. Hence, significantly less effort is needed to improve the technological parameters of the production process.