Limit theorems for nondegenerate \(U\)-statistics of continuous semimartingales (Q473167)

For a sequence of observations from a one-dimensional continuous Itō semimartingale, a U-statistic is defined by extending the similar definition based on a sequence of independent and identically distributed observations. The uniform convergence in probability (i.e., the law of large numbers) of this \(U\)-statistic is first proved. Then, a stable central limit theorem is established. With an estimate for the conditional variance, a standard central limit theorem is also proved for a standardized version of the \(U\)-statistic. These general results are applied to derive the limit distributions for Gini's mean difference, which may be viewed as an alternative measure of price variability in mathematical finance, an \(L^p\)-type test statistic for constant volatility, and the Wilcoxon test statistic for structure breaks.