Spectral analysis for intrinsic time processes

This paper discusses an estimation procedure for the spectral density of intrinsic time processes because there has been no argument of the spectral analysis for subordinated processes. Such processes have been proposed in a variety of contexts to describe asset price behavior. They are used when the movement of prices is tied to the number of market transactions, trading volume or the more illusive concept of information arrival. We develop the asymptotic theory for an estimated spectral density of intrinsic time processes and elucidate the asymptotics, which show some interesting structures. Also, numerical studies are given to confirm the results.