{"id":"arxiv/2610.39177","url":"https://arxiv.org/abs/2610.39177","tags":["math.ST"],"score":0,"title":"Adaptive Bayesian Estimation of Conditional Densities Based on Orthogonal Polynomials","domain":"math.ST","source":"arxiv","authors":["Andriy Norets","Marco Stenborg Petterson"],"reviews":0,"subject":"math.ST","summary":"We study Bayesian estimation of conditional densities represented by normalized squares of orthogonal polynomial series. We specify the conditional density without modeling the marginal distribution of covariates. A prior on smoothness, the number of terms, and the coefficients gives adaptive posterior contraction at the minimax rate without a logarithmic factor when the normalized square-root representation of the true density is bounded away from zero and sufficiently smooth. For one outcome and $d_x$ covariates, the rate over classes with degree smoothness $s$ is $n^{-s/(2s+d_x+1)}$. The…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-29","reactions":{}}