McAssey, M.P.

For personal home page click here

Research statement

I have recently been working with primatologists at the University of California to develop a method for computing a nonlinear social hierarchy for captive rhesus macaques based on conflict data.  I have adapted this method for use in ranking sports teams.  Previous projects include the development of an algorithm for computing the number of clusters in a data cloud---along with cluster membership, estimation of the slope in heteroscedastic measurement error models without equation error, modeling instantaneous coupling states among EEG gamma signals in a rat's brain using a mixed multivariate beta distribution, and computing conditions for coherence in a network equipped with neuron-like signal transmissions as well as estimating ideal edge-probabilties to optimize the likelihood of network coherence when the edges are random.  My work has involved development of efficient code in R to carry out data analyses, simulations, and optimization routines such as the simulated annealing and expectation-maximization algorithms.  I have enjoyed collaboration with researchers from a variety of disciplines, including psychology, neuroscience, epidemiology and wildlife biology, and am now working with scientists in the Computational Neuroscience Group at VU Amsterdam.