Moment estimation in discrete shifting level model applied to fast array-CGH segmentation

We develop a mathematical theory needed for moment estimation of the parameters in a general shifting level process (SLP) treating, in particular, the finite state space case geometric finite normal (GFN) SLP. For the SLP, we give expressions for the moment estimators together with asymptotic (co)variances, following, completing, and correcting cline (Journal of Applied Probability 20, 1983, 322–337); formulae are then made more explicit for the GFN-SLP. To illustrate the potential uses, we then apply the moment estimation method to a GFN-SLP model of array comparative genomic hybridization data. We obtain encouraging results in the sense that a segmentation based on the estimated parameters turns out to be faster than with other currently available methods, while being comparable in terms of sensitivity and specificity.