{"id":"arxiv/2610.05457","url":"https://arxiv.org/abs/2610.05457","tags":["math.ST","math.PR"],"score":0,"title":"Measuring the roughness of a signal","domain":"math.ST","source":"arxiv","authors":["Rama Cont","Purba Das"],"reviews":0,"subject":"math.ST","summary":"We study the non-parametric estimation of the roughness of a path from discretely sampled observations. Our approach is based on the concept of normalized power variation of a path. The estimator identifies a variation index by comparing coarse increments with local sums of fine increments. We prove pathwise consistency under Hölder regularity, nondegenerate coarse critical power sums, uniform local bounds on fine critical power sums, and separation of the two sampling scales and obtain a convergence rate for the estimator. Neither convergence of critical variation nor monotonicity of the…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-04","reactions":{}}