Smoothing and Ill-Posed Problems
The method of weighted cross-validation is applied to the problem of solving linear integral equations of the first kind with noisy data. Numerical results illustrating its efficacy are given for estimating derivatives and for solving Fujita’s equation.
KeywordsRidge Regression Reproduce Kernel Hilbert Space Subset Selection Linear Integral Equation Principal Component Method
Unable to display preview. Download preview PDF.
- 1.Gehatia, M., and Wife, D. R., Solution of Fujita’s Equation for Equilibrium Sedimentation by Applying Tikhonov’s Regularizing Functions, Journal of Polymer Science Series A2, Vol. 8, pp. 2039–2050, 1970.Google Scholar
- 3.Mallows, C. L., Some Comments on Cp, Technometrics, Vol. 15, pp. 661675, 1973.Google Scholar
- 4.Hudson, H. M., Empirical Bayes Estimation, Department of Statistics, Stanford University, Technical Report No. 58, 1974.Google Scholar
- 5.Craven, P., and Wahba, G., Smoothing Noisy Data with Spline Functions: Estimating the Correct Degree of Smoothing by the Method of Generalized Cross-Validation, Numerische Mathematik, Vol. 31, pp. 377–403, 1979. Report No. 445, 1977.Google Scholar
- 6.Golub, G., Heath, M., and Wahba, G., Generalized Cross-Validation as a Method for Choosing a Good Ridge Parameter, Department of Statistics, University of Wisconsin at Madison, Technical Report No. 491, 1977.Google Scholar
- 9.Stone, M., Cross-Validatory Choice and Assessment of Statistical Prediction, Journal of the Royal Statistical Society Series B, Vol. 36, pp. 111–147, 1974.Google Scholar
- 12.Varah, J. E., A Practical Examination of Numerical Methods for Linear Discrete Ill-Posed Problems, Department of Computer Sciences, University of British Columbia, Technical Report No. 76–08, 1976.Google Scholar