Abstract
In this review, we introduce the framework of Bayesian estimation in spectral deconvolution. In Bayesian estimation, we can trace the causal relationships by Bayes’ theorem to extract the underlying structure behind the spectral data. By applying Bayesian estimation to spectral deconvolution, we can estimate the number of peaks in the spectral data based on the framework of model selection. Furthermore, by using an algorithm called exchange Monte Carlo method, it is possible to solve the problem that is trapped in the local optimum solution.
