A1295
Title: Intraday correlation patterns
Authors: Zhiyuan Zhang - Shanghai University of Finance and Economics (China) [presenting]
Torben G Andersen - Kellogg School, Northwestern University (United States)
Viktor Todorov - Northwestern University (United States)
Yingwen Tan - HKUST (China)
Abstract: A functional central limit theorem is established for estimating the diurnal pattern of instantaneous correlations between two financial assets, using high frequency data over a long span of time. This is the first functional central limit theory ever built for such high frequency characteristics of asset return correlations. An estimator of the asymptotic covariance function is proposed, rendering the limit theorem feasible in a two-sample test for the equivalence of diurnal patterns of two non-overlapping time periods. Simulation evidence supports the theoretical findings, while empirical results on the E-mini S\&P 500 and US Treasury bond futures show that correlations exhibit pronounced intraday periodicity, with the diurnal patterns varying significantly over time.
