2010/10/16 by Ronald P. Barry, Barry, Ronald P., Julie McIntyre +1
Environmental Science · #Applications (stat.AP) #Avian ecology and behavior #FOS: Biological sciences #FOS: Computer and information sciences #Fish Ecology and Management Studies #Methodology (stat.ME) #Quantitative Methods (q-bio.QM) #Remote Sensing and LiDAR Applications
paper · pdf · doi:10.48550/arxiv.1010.3317
openalex publication_date 2010/10/16 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28
Density estimates based on point processes are often restrained to regions\nwith irregular boundaries or holes. We propose a density estimator, the\nlattice-based density estimator, which produces reasonable density estimates\nunder these circumstances. The estimation process starts with overlaying the\nregion with nodes, linking these together in a lattice and then computing the\ndensity of random walks of length k on the lattice. We use an approximation to\nthe unbiased crossvalidation criterion to find the optimal walk length k. The\ntechnique is illustrated using walleye (Sander vitreus) radiotelemetry\nrelocations in Lake Monroe, Indiana. We also use simulation to compare the\ntechnique to the traditional kernel density estimate in the situation where\nthere are no significant boundary effects.\n