Progress and problems in estimating climage variables and covariances from incomplete data sets

Tapio Schneider
California Institute of Technology

Abstract
I review statistical issues arising in the estimation of climate variables and covariances from incomplete datasets, including the incomplete datasets consisting of a mixture of climate proxies and instrumental climate data that occur in climate reconstruction problems. Particular emphasis will be placed on the problems arising from rank deficiency and ill-posedness in estimation procedures. Several estimation procedures and regularization strategies and their relative merits will be discussed.

Back to agenda