Hi, I am approximation a value function (in dynamic programming) V: R^n->R. In different versions of the problem, n is between 2 and 5. I would like to use splines to use this, I checked the mailing list and CRAN but couldn't find anything that would help me (Tps in fields is too slow, cobs would be really nice but seems to be one-dimensional). Is there any package that would at least give me the basis for multidimensional b-splines? Or even one that would do the fitting, perhaps even with constaints on derivatives at the edges (which I can specify from theory)? Thank you, Tamas
Bayesian statistics is difficult in the sense that thinking is difficult. --Donald A. Berry, American Statistician 51:242 (1997)