Choice of smoothing parameters in wavelet series estimators
Geunghee Lee · Journal of nonparametric statistics · 2003
Wavelets have received considerable interest in denoising since Donoho and Johnstone (1994 Donoho, D. and Johnstone, I. (1994). Ideal spatial adaptation by wavelet shrinkage. Biometrika, 81: 425–455. [Crossref], [Web of Science ®] , [Google Scholar]) introduced wavelet series estimators. The performance of wavelet series estimators depends on two smoothing parameters – a thresholding starting level j 0 and a threshold δ. We derive a data criterion to choose two smoothing parameters (j 0 and δ) simultaneously in wavelet series estimators.