Zhang, XibinKing, MaxwellShang, Hanlin2015-12-070167-9473http://hdl.handle.net/1885/23388The unknown error density of a nonparametric regression model is approximated by a mixture of Gaussian densities with means being the individual error realizations and variance a constant parameter. Such a mixture density has the form of a kernel densityKeywords: Learning algorithms; Mathematical models; Mixtures; Regression analysis; Value engineering; Bayes factor; Error density; Metropolis-Hastings algorithm; Predictive density; Value at Risk; Bandwidth Bayes factors; Kernel-form error density; Metropolis-Hastings algorithm; Posterior predictive density; State-price density; Value-at-risk201410.1016/j.csda.2014.04.0162019-08-18