|
|
Royal Statistical Society |
| 1997 | |
|---|---|
|
2.30 p.m., 15 October |
JAMES S. HODGES (University of Minnesota)
Some algebra and geometry for hierarchical models, applied to diagnostics Hierarchical linear models are popular and readily computed. Diagnostic methods for such models are lagging behind practice, in part because we have no unifying theory, as we have for single-level linear models. This paper proposes a unifying theory, derives several diagnostics, and applies them to a problem. Appears (with discussion) in Journal of the Royal Statistical Society, Series B, 60(3), 497-536 (1998). |
| 1998 | |
|
5.00 p.m., 20 May |
STEVE DAVIES and PETER HALL (ANU)
Fractal analysis of surface roughness using spatial data Fractal methods for analysing surface roughness, using pixelated data on surface height, are introduced and developed. Particular emphasis is given to the issue of anisotropy. It is shown that, for many surfaces, the fractal dimension of line transects must either be constant in every direction, or be constant in each direction except one. Practical implications of this property are addressed. Data on the surfaces of soil and plastic food wrap are used to illustrate applications of our methodology. Appears (with discussion) in Journal of the Royal Statistical Society, Series B, 61(1), 3-37 (1999). |