Motivated by features of low latency data in finance we study in detail discrete-valued Levy processes as the basis of price processes for high frequency econometrics. An important case of this is a Skellam process, which is the difference of two independent Poisson processes. We propose a natural generalisation which is the difference of two negative binomial processes. We apply these models in practice to low latency data for a variety of different types of futures contracts.
Keywords:
low latency data
,high frequency econometrics
,Skellam distribution
,futures markets
,negative binomial