PROJECTED MULTIVARIATE LINEAR-MODELS FOR DIRECTIONAL-DATA

We introduce the spherically projected multivariate linear modal for d irectional data. This model treats directional observations as project ions onto the unit sphere of unobserved responses from a multivariate linear model. Focusing on the important case of circular data, we show that maximum likelihood estimates for the model are readily computed using iterative methods, in sharp contrast with competing approaches. Examples are given to demonstrate the resulting methodology in realist ic applications.