Prior Selection By An Adaptive Smoothing Splines Approach

Under the context of empirical bayes a prior density estimate is obtained by using B-splines. In this approach, there are two smoothing parameters, the number of basis functions and usual regularization parameter found in the context of penalized least squares problem. An algorithm is provided to get estimates of the two parameter as well as the prior density. © 1998 VSP.615760Bennet, G.K., Martz, H.F., A continuous empirical bayes smoothing technique (1972) Biometrica, 50, pp. 361-368Berger, J.O., (1985) Statistical Decision Theory and Bayesian Analysis, , (Second Edition), Springer-VerlagDe Boor, C., (1978) A Practical Guide to Splines, , Springer Verlag, New YorkDias, R., (1996) Sequential Adaptive Nonparametric Regression Via Splines, , Technical Report RP 43/96, University of Campinas, June SubmittedDias, R., Density estimation via splines Journal of Statistical Computation and Simulation, , To appearGu, C., Smoothing spline density estimation: A dimensionless automatic algorithm (1993) J. of the Amer. Statl Assn., 88, pp. 495-504Kooperberg, C., Stone, C.J., A study of logspline density estimation (1991) Computational Stat. and Data Analysis, 12, pp. 327-34