The Log-Periodic Power Law Model: An Exploration
Date
2016-08
Authors
Contributor
Advisor
Department
Instructor
Depositor
Speaker
Researcher
Consultant
Interviewer
Narrator
Transcriber
Annotator
Journal Title
Journal ISSN
Volume Title
Publisher
[Honolulu] : [University of Hawaii at Manoa], [August 2016]
Volume
Number/Issue
Starting Page
Ending Page
Alternative Title
Abstract
Over the last two decades, a new financial model has emerged that might explain some of the rare instances of truly extreme volatility in asset prices. This model, known as the Log-Periodic Power Law (LPPL) model or the Johansen-Ledoit-Sornette (JLS) model, attempts to diagnose, time, and predict the termination of these bubbles; we caution that there is no academic agreement about the existence or definition of a “bubble.” The creators of the model provide a motivation built upon some natural assumptions including risk-neutral assets, rational expectations, local self-reinforcing imitation, and probabilistic critical times. The model has evolved over time, partially in response to some sound criticism. This dissertation is focused on two criticisms that have not been fully addressed. First, it is unknown whether there exist unique best fits of the JLS model to log-price data. In this dissertation we explore the first level JLS model and analyze its relationship with extreme boundary vectors which serve as the building blocks for increasing concave up log-price paths. Second, it is unknown whether the current method for locating local minima is sufficient. Using numerical analysis, this dissertation uses the Cauchy-Schwarz Theorem and Taylor’s Theorem to find bounds for various moduli of continuity and first and second derivatives of the error of the JLS model fit to appropriate log-price paths. In addition to discussing these criticisms, the question of the applicability of the JLS model is considered. Without committing ourselves to a definition of a bubble, this dissertation also presents the results of an ongoing test attempting to determine whether the JLS model can be used to generate systematic profits. An experiment using the JLS model on specifically chosen stocks filtered from the New York Stock Exchange (NYSE) is detailed along with a step-by-step procedure to determine specific dates on which to utilize a trading strategy.
Description
Ph.D. University of Hawaii at Manoa 2016.
Includes bibliographical references.
Includes bibliographical references.
Keywords
Citation
Extent
Format
Geographic Location
Time Period
Related To
Theses for the degree of Doctor of Philosophy (University of Hawaii at Manoa). Math
Related To (URI)
Table of Contents
Rights
Rights Holder
Local Contexts
Collections
Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.