Date of Award

1-1-2014

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

College/School/Department

Department of Physics

Content Description

1 online resource (xii, 198 pages) : illustrations (some color)

Dissertation/Thesis Chair

Kevin H. Knuth

Committee Members

Ariel Caticha, Carolyn MacDonald, Jesse Ernst, Amos Golan

Keywords

Computational Physics, Crashes, Log Periodic Oscillation, Power Law, Stock Indices, Stock Market, Stocks, Financial crises, Econophysics, Statistical power analysis, Statistical physics

Subject Categories

Physics

Abstract

Viewing the stock market as a self-organized system, Sornette and Johansen introduced physics-based models to study the dynamics of stock market crashes from the perspective of complex systems. This involved modeling stock market Indices using a mathematical power law exhibiting log-periodicity as the system approaches a market crash, which acts like a critical point in a thermodynamic system. In this dissertation, I aim to investigate stock indices to determine whether or not they exhibit log-periodic oscillations, according to the models proposed by Sornette, as they approach a crash. In addition to analyzing stock market crashes in the frequency domain using the discrete Fourier transform and the Lomb-Scargle periodogram, I perform a detailed analysis of the stock market crash models through parameter estimation and model testing. I find that the probability landscapes have a complex topography and that there is very little evidence that these phase transition-based models accurately describe stock market crashes.

Download

Included in

Physics Commons

Share

COinS