{"id":"arxiv/2609.13553","url":"https://arxiv.org/abs/2609.13553","tags":["stat.ME","math.PR"],"score":0,"title":"A Conditional-Distribution Framework for Validating Synthetic Multivariate Data","domain":"stat.ME","source":"arxiv","authors":["Hari Dahal","Ishanu Chattopadhyay"],"reviews":0,"subject":"stat.ME","summary":"Statistical validation of synthetic multivariate data requires assessing whether a generator preserves the joint dependence structure of the target population without merely reproducing observed records. We develop a model-agnostic framework based on full conditional distributions. For each coordinate, we normalize the conditional probability assigned to the observed value by the largest conditional probability available in the same record context; averaging this quantity yields a one-sided MAP-alignment statistic that can be estimated using a conditional model fitted on held-out real data.…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-09-11","reactions":{}}