Deconvolution Density Estimation
We review the scheme introduced by Kim and Richards for deconvolution density estimation on compact Lie groups. The Fourier transform is used to express the problem in terms of products of matrices of irreducible representations. We introduce a sequence of consistent estimators, by taking cut-offs in a certain Fourier expansion. Some smoothness classes of noise are discussed and used to find optimal rates of convergence of the estimators.