荷世平臺灣大學：土木工程學研究所梁佩芬Leong, Pui FanPui FanLeong2007-11-252018-07-092007-11-252018-07-092006http://ntur.lib.ntu.edu.tw//handle/246246/50197The recent bull markets around the world have led many academics and practitioners to predict if there exist asset price bubbles. To judge whether there is an asset price bubble, economist usually use a model of the fundamental value to check the existence of bubbles. Since the fundamental value is related to the future expectation and uncertainty of the market, economists sometimes define bubbles differently. Due to the difficulties associated with the assessment of the fundamental value, one can hardly detect bubbles using a standard model. Therefore, our main question in this study is “What is the difference between bubbles and regular fluctuations (market fundamental value) of stock markets and real estate markets?” Using the Augmented Dickey-Fuller (ADF) unit root test, we find that the stock market indexes are non-stationary. Furthermore, many empirical results suggest that the economic behavior, like investors’ attitudes toward risk and expected return, are nonlinear. The Hilbert-Huang Transform (HHT) which is developed by Huang et al. (1998) is a new method that can better study both nonlinear and nonstationary data. Using HHT, this paper seeks to answer whether there is a significant difference between market fundamentals and speculative bubbles and how to identify the signals that can detect the formation and crash of economic bubbles. A valid signal for detecting bubbles can help investors or government monitor the health condition and anomalies of real estate and financial markets and prevent major economic losses.Abstract i Table of Contents ii List of Tables iv List of Figures v Chapter 1 Introduction 1 1.1 Motivation and Objective 1 1.2 Organization of the Thesis 2 Chapter 2 Literature Reviews 3 2.1 Theoretical Background of Bubbles and Noise Trading 3 2.1.1 Rational Bubbles 3 2.1.2 Irrational Bubbles 5 2.1.3 Noise Trading 6 2.2 Financial applications of HHT 8 Chapter 3 Methodology 10 3.1 Testing for Nonstationarity 10 3.1.1 Dickey-Fuller Test 10 3.1.2 Augmented Dickey-Fuller Test 11 3.2 Hilbert Huang Transform 12 3.2.1 Empirical Mode Decomposition 12 3.2.2 Hilbert Spectral Analysis 17 Chapter 4 Empirical Results and Analyses 19 4.1 Data Description 19 4.2 The Empirical Results of Unit Root Test 20 4.3 US Stock Internet Bubble in 1990s 21 4.3.1 Introduction 21 4.3.2 Cases Description 22 4.3.3 Empirical Results of HHT in US Stock Market 23 4.3.4 Conclusions 35 4.4 Japan Stock Market in 1980s 36 4.4.1 Introduction 36 4.4.2 Case Description 36 4.4.3 Empirical Results of HHT in Japan Stock Market 37 4.4.4 Conclusions 40 4.5 Taiwan Stock Market in 1980s 41 4.5.1 Introduction 41 4.5.2 Case Description 41 4.5.3 Empirical Results of HHT in Taiwan Stock Market 42 4.5.4 Conclusions 45 4.6 Hong Kong Real Estate Market in 1990s 46 4.6.1 Introduction 46 4.6.2 Case Description 46 4.6.3 Empirical Results of HHT in Hong Kong Real Estate Market 47 4.6.4 Conclusions 50 Chapter 5 Conclusions and Suggestions 51 5.1 Conclusions 51 5.2 Further Studies 52 Bibliography 531450448 bytesapplication/pdfen-US希爾伯特-黃轉換經驗模態希爾伯特頻譜分析泡沫現象Hilbert-Huang TransformEMDHSAbubblesthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/50197/1/ntu-95-R93521705-1.pdf