A cross validated Bayesian retrieval algorithm for nonlinear remote sensing experiments (Q580903)

The paper extends the method of generalized cross-validation to nonlinear problems of the form \(Y_ i=N(\theta,x_ i)+\epsilon_ i,\) \(i=1,...,n\). Here, \(Y_ i\) are the measurements, \(x_ i\) the design points, \(\epsilon_ i\) measurement errors, and \(\theta\) the parameter to be estimated. An application to a remote sensing problem of atmospheric temperature profiles is described in detail.