{"id":"arxiv/2605.13504","url":"https://arxiv.org/abs/2605.13504","tags":["stat.ME","math.AP","math.DS","math.PR","q-bio.QM"],"score":0,"title":"Structural identifiability of partially-observed stochastic processes: from single-particle trajectories to total particle density data","domain":"stat.ME","source":"arxiv","authors":["Arianna Ceccarelli","Alexander P. Browning","Ruth E. Baker"],"reviews":0,"subject":"stat.ME","summary":"The increasing availability of experimental data has intensified interest in calibrating stochastic models, raising fundamental questions about parameter identifiability. Structural identifiability determines whether parameters can be uniquely recovered from idealised, noise-free data, a prerequisite to allow for parameter estimation in real-world scenarios. However, existing methods to assess structural identifiability are not generally applicable to stochastic processes. We develop a methodology to analyse structural identifiability for a class of stochastic processes. We investigate how…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-05-13","reactions":{}}