As a tradeoff between efficiency and costs \mbox{modern} communication systems contain a variety of components that can at least be considered weakly nonlinear. A critical element in evaluating the degree of nonlinearity of any underlying nonlinear system is the amount of undesired signal strength or signal power this system is introducing outside the transmission bandwidth. This phenomenon called spectral regrowth or spectral broadening is subject to stringent restrictions mainly imposed by the given specifications of the particular communication standard. Consequently, achieving the highest possible efficiency without exceeding the linearity requirements is one of the main tasks in system design. Starting from this challenging engineering problem there grows a certain need for specialized tools that are capable of predicting linearity and efficiency of the underlying design. Besides a multitude of methods aiming at the prediction of spectral regrowth a statistical approach in modeling and analyzing nonlinear systems offers the advantage of short processing times due to closed form mathematical expressions in terms of input and output power spectra and is therefore further examined throughout this article.