Testing for the Markov Property in a High-Frequency Setting

Testing for the Markov Property in a High-Frequency Setting (joint with Yacine Ait-Sahalia)

The aim is to present a test for the homogeneous Markov property of a one-dimensional process
X observed at regularly spaced times over a finite time interval. The frequency goes to infinity, and
we test the null hypothesis according to which the spot volatility takes the form σt = f(Xt) for some
smooth enough non-vanishing function f. The test relies on some Central Limit Theorems related
to the local times of a semimartingale. We allow the process X to have jumps, restricted to finite
activity. We will mostly consider the case when the process is observed without error, and if time
permits we will give a method covering the case where microstrucutre noise is present.
 
Zoom Link: https://univr.zoom.us/j/88333677482