Modelling the Bivariate Spatial Distribution of Amacrine Cells
We are interested in studying the spatial dependency between the positions of on and off cholinergic amacrine cells because we hope this will tell us something about how the two cell types emerge during development.
Our goal in is to demonstrate how recently developed Monte Carlo methods for conducting likelihood-based analysis of realistic point process models can lead to sharper inferences about the bivariate structure of the data. In particular, we will formulate and fit a bivariate pairwise interaction model for the amacrines data, and will argue that likelihood-based inference within this model is both statistically more efficient and scientifically more relevant than ad hoc testing of benchmark hypotheses such as independence.
Key wordsBivariate pairwise interaction processes Cholinergic amacrine cells Likelihood-based inference Monte-Carlo methods
Unable to display preview. Download preview PDF.
- P.J. Diggle. Statistical Analysis of Spatial Point Patterns, 2nd ed. Oxford University Press, 2003.Google Scholar
- E.V. Famiglietti. Starburst amacrine cells: morphological constancy and systematic variation in the anisotropic field of rabbit retinal neurons. Journal of Neuroscience, 5:562–577, 1985.Google Scholar
- C.J. Geyer. Likelihood inference for spatial point processes. In O.E. Barndorff-Nielsen, W.S. Kendall and M.N.M. van Lieshout, editors, Stochastic Geometry: likelihood and computation, pages 141–172. Chapman and Hall/CRC, 1999.Google Scholar
- M.N.M. van Lieshout. Markov point processes and their applications. Imperial College Press, 2000.Google Scholar
- E. Wienawa-Narkiewicz. Light and Electron Microscopic Studies of Retinal Organisation. PhD thesis, Australian National University, Canberra, 1983.Google Scholar