Date of Award

Spring 5-15-2015

Author's School

Graduate School of Arts and Sciences

Author's Department


Degree Name

Doctor of Philosophy (PhD)

Degree Type



This dissertation proposes a methodology for inference in the context of diffusion processes with jumps. There are many applications. For example, in finance, this methodology can be used to study asset pricing. My dissertation consists of two chapters which are closely related. They reveal the relationship between the power of a test, jump height and jump frequency. In the first chapter I construct a likelihood ratio test to test whether a diffusion process has jumps. This test statistic is independent of the distribution of jump height. I show the test is asymptotically optimal when the jump height is O(1/n^alpha) , the jump frequency is O(1/n^beta) where n is sample size, 3alpha+beta=2,alpha>1/2,beta>0. By constructing this optimal test, I derive the asymptotic power envelopes for testing continuous diffusion process against diffusion processes with asymmetric jumps. In recent years, many tests for this problem were proposed. I compare the power of these tests with the envelopes using simulations.

In chapter two I test the continuous diffusion process against a diffusion process with symmetric jumps. I show my test statistic is optimal when the jump height is O(1/n^alpha), the jump frequency is O(1/n^beta) where n is sample size, 4alpha+beta=5/2, alpha>2/3, 0


English (en)

Chair and Committee

Werner Ploberger

Committee Members

George-Levi Gayle, Nan Lin, John Nachbar, Jonathan Weinstein


Permanent URL:

Available for download on Wednesday, May 15, 2115

Included in

Economics Commons