Bates, Ron A., Curtis, Peter R., Maruri-Aguilar, Hugo and Wynn, Henry P. (2013) Optimal design for smooth supersaturated models. Journal of Statistical Planning and Inference . ISSN 0378-3758 (In Press)
Smooth supersaturated models are interpolation models in which the underlying model size, and typically the degree, is higher than would normally be used in statistics, but where the extra degrees of freedom are used to make the model smooth using a standard second derivative measure of smoothness. Here, the solution is derived from a closed-form quadratic programme, leading to tractable matrix representations. This representation aids considerably in the choice of optimal knots in the interpolation case and in the optimal design when the SSM is used as a way of obtaining kernels, but where the statistical problem is set up separately. Some examples are given in one and two dimensions.
|Additional Information:||© 2013 Elsevier B.V.|
|Library of Congress subject classification:||H Social Sciences > H Social Sciences (General)|
|Sets:||Research centres and groups > Decision Support and Risk Group (DSRG)|
|Date Deposited:||06 Jan 2014 11:56|
Actions (login required)
|Record administration - authorised staff only|