proxymix
Fits multivariate Gaussian-mixture proxies to target distributions on real Euclidean space, given by a density, by samples, or by a characteristic function. Four fitting regimes are unified under one verb: (i) closed-form moment matching for a single component, (ii) classical expectation-maximisation when independent samples are available, (iii) importance-sampled expectation-maximisation that minimises the Kullback-Leibler divergence when the target can be evaluated point-wise but not (cheaply) sampled, and (iv) a least-squares fit to the characteristic function when only that is known, with numerical bounds on the Kolmogorov distance along one-dimensional projections. Closed-form Gaussian-mixture operators (density, sampling, marginalisation, conditioning, divergence) round out the toolkit. The conditioning operator drives multiple imputation of data missing at random, covering the multimodal and heteroscedastic cases a single-Gaussian model cannot represent. Implements the regime hierarchy of van der Hoek and Elliott (2024) <doi:10.1080/07362994.2024.2372605>.
This page was built for software: proxymix
