[[alternative]]影響股票市場報酬的驅動因素 : 應用縱橫平滑移轉模型的實證探討

博士[[abstract]]由於近年來，全球經濟一直受到油價波動、全球金融風暴與美國量化寬鬆政策的影響，因此股票市場在遭受不同風險情況下，對於報酬的影響應該也會不同。故本文擬加入不同的風險因子(油價、全球市場與匯率風險)，並運用縱橫平滑移轉迴歸模型(Panel Smooth Transition Regression)分別來探討股票市場報酬如何受到總體經濟、財務理論與國際金融相關因素的影響。 本文第一個研究議題為「在油價波動的衝擊下，探討總體經濟與股票市場報酬的非線性關係」，實證結果發現：當油價風險較小時(約小於5%)，貨幣供給(Money Supply)對於股票市場報酬影響較大，反之，當油價風險較大時(約大於5%)，則是實質經濟活動(Real Activity)對於股票市場報酬影響較大。 第二個研究議題為「探討資本資產定價模型的非線性關係」，本篇利用全球市場風險做為門檻變數，探討 與股票市場報酬的非線性關係，實證結果發現： 與股票市場報酬存在正向關係；但是當全球市場風險較小時，反而對於股票市場報酬的影響較為顯著。 本文最後的研究議題為「探討在匯率風險下，外人投資與股票市場報酬的非線性關係」，實證結果發現：在匯率風險較小時，不論外人投資的型態為何，對於股票市場報酬都為正向的關係；然而當匯率風險較大時，外人直接投資(Foreign Direct Investment)與外人證券投資(Foreign Portfolio Investment) 對於股票市場報酬則為負向的關係。[[abstract]]In recent years, oil price volatility, global financial crises and America''s Quantitative Easing Monetary Policy have been affecting the global economy. Hence, stock market returns should be affected based on different investment risks. For this reason, the purpose of this dissertation utilize the Panel Smooth Transition Regression (PSTR) model to investigate the influence of the stock market returns on macroeconomics, financial theory and international finance from three different risk factors (oil, global market and exchange rate risks). The topic of the first part is: Nonlinear investigation for the impact of oil price volatility on the fundamental analysis. Empirical evidence shows that when real oil price volatility is in a low regime (approximately less than 5%), money supply tends to have a greater effect on real stock returns. Conversely, when real oil price volatility is in a high regime (approximately greater than 5%), real activity have a greater effect on real stock returns. The topic of the second part is: Threshold effects in the capital asset pricing model. We set the volatility of world market excess return as the threshold variable, and to examine the nonlinear relationship between beta (systematic risk) and returns (world market excess returns). The results show that all beta values are positive and higher in the low regime (i.e., volatility of world market excess return is low) and lower in the high regime (i.e., volatility of world market excess return is high). Finally, The topic of the third part is: Nonlinear relationship between foreign investments and stock market returns with exchange rate volatility. The results show that all foreign investments are positive in the low regime for exchange rate volatility, , no matter what kind of investment. But in the high regime, the relationship between the stock market returns both foreign direct and portfolio investments have negative relations.[[tableofcontents]]Table of Content Chapter 1 Introduction 1 Chapter 2 Literature Review 4 2.1 Relationship between macroeconomic variables and stock market returns 4 2.2 Issues about Capital Asset Pricing Model (CAPM) 6 2.3 Relationship between foreign investments and stock market returns 8 Chapter 3 Methodology 11 3.1 Panel Unit Root Test 11 3.2 Panel Smooth Transition Regression (PSTR) 13 Chapter 4 Empirical Investigation 21 4.1 Nonlinear Investigation for the Impact of Oil Price Volatility on the Fundamental Analysis 21 4.1.1 Data and Summary Statistics 21 4.1.2 Empirical results 23 4.2 Threshold effects in the capital asset pricing model using panel smooth transition regression (PSTR) Evidence from net oil export and import groups 29 4.2.1 Data and Summary Statistics 29 4.2.2 Empirical results 32 4.3 Nonlinear relationship between foreign investments and stock market returns with exchange rate volatility 37 4.3.1 Data and Summary Statistics 37 4.3.2 Empirical results 39 Chapter 5 Conclusions 43 References 47 List of Tables Table 4.1.1 Countries and stock price indices employed in the study. 22 Table 4.1.2 Descriptive statistics 22 Table 4.1 3 Panel unit root tests 23 Table 4.1 4 LM test for remaining nonlinearity 24 Table 4.1 5 Determination of the number of location parameters 25 Table 4.1 6 Parameter estimates for the final PSTR models 26 Table 4.1 7 Estimation of coefficients of control variables in PSTR models 26 Table 4.2.1 Stock price indices and risk-free rate employed in the study. 30 Table 4.2.2 Summary statistics 31 Table 4.2.3 Panel unit root tests 32 Table 4.2.4 LM test for remaining nonlinearity 33 Table 4.2.5 Determination of the number of location parameters 33 Table 4.2.6 Parameter estimates for the final PSTR models 35 Table 4.2.7 Estimation of coefficients of control variables in PSTR models 35 Table4.3.1 Summary statistics 38 Table4.3.2 Panel unit root tests 39 Table4.3.3 LM test for remaining nonlinearity 40 Table4.3.4 Determination of the number of location parameters 40 Table4.3.5 Parameter estimates for the final PSTR models 41 Table4.3.6 Estimation of coefficients of control variables in PSTR models 42 List of Figures Figure 4.1.1 Transition function with respect to transition variable for model A. 28 Figure 4.1.2 Transition function with respect to transition variable for model B. 28[[note]]學號: 895530151, 學年度: 10