Hostname: page-component-7b7567c8c-7pwfn Total loading time: 0 Render date: 2026-11-16T11:14:01.892Z Has data issue: false hasContentIssue false

CONDITIONS FOR THE PROPAGATION OF MEMORYPARAMETER FROM DURATIONS TO COUNTS AND REALIZEDVOLATILITY

Published online by Cambridge University Press:  01 June 2009

Rights & Permissions [Opens in a new window]

Abstract

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the 'Save PDF' action button.

We establish sufficient conditions on durations thatare stationary with finite variance and memoryparameter $d
                                                  \in [0,{\textstyle{1 \over
                                                  2}})$ to ensure thatthe corresponding counting processN(t) satisfiesVar N(t) ~Ct2d+1(C > 0) as t→ ∞, with the same memory parameter $d
                                                  \in [0,{\textstyle{1 \over
                                                  2}})$ that was assumedfor the durations. Thus, these conditions ensurethat the memory parameter in durations propagates tothe same memory parameter in the counts. We thenshow that any autoregressive conditional durationACD(1,1) model with a sufficient number of finitemoments yields short memory in counts, whereas anylong memory stochastic duration model withd > 0 and all finite momentsyields long memory in counts, with the samed. Finally, we provide someresults about the propagation of long memory to theempirically relevant case of realized varianceestimates affected by market microstructure noisecontamination.

Information

Type
Research Article
Copyright
Copyright © Cambridge University Press 2009