Skip to Main Content
The development and comparison of a class of nonparametric probability density function modeling algorithms is presented. Each algorithm Iteratively estimates a model of the sampled density function based upon a description of a set of equiprobable regions over the range of the variable of interest. Minimization of computational complexity and memory capacity while maintaining convergence and stability are the principal considerations. Variations of the algorithms are compared as to rate of convergence and limit cycle stability relative to ease of implementation. Results including comparative curves are presented.