An Improved Swarm Intelligence Based Method for Particle Size Change in Optimization Process

The Swarm Intelligence Based (SIB) method received much attention from statisticians and researchers in related fields in the past few years for its ability to efficiently search for optimal results in discrete domains like problems in designs of experiments. However, it was still of limited use because the change of particle size was forbidden during the optimization process. In this paper, we introduced a new VARY operation to achieve particle size change during the optimization process. We applied the improved method to a classic statistical problem in finding change points on functional data to achieve optimal model fitting.