[[alternative]]黃金價格、石油價格、匯率和消費支出的門檻關係

博士[[abstract]]自2008年的金融海嘯後，黃金市場吸引著來自全球的投資者投入資金進行避險，當投資者進行投資組合規劃時，美元兌歐元的匯率變動和油價變化也深深影響著全球資金的流動和金融市場。因此在此篇論文的第一個研究議題上，以門檻效果的STAR模型來探討黃金價格、油價及匯率的非線性關係，實證結果發現以匯率做為門檻變數時，三個變數的關係為LSTAR模型，而以油價為門檻變數時，則是ESTAR模型，不論是在LSTAR和ESTAR模型，金價與匯率均存在著較大的門檻效果。 而在第二和第三的研究議題上，我們採用縱橫平滑轉換模型(PSTR)來分析石油出口國及進口國在油價與私人消費上的門檻效果，因為考量到消費者在油價上漲時會產生遞延消費的現象，因此以落後一期的油價做為門檻變數。實證結果顯示，當門檻變數落入低和高區間時，石油出口國的消費者會減少其消費支出；相反地，在石油進口國的實證結果中發現，當門檻變數落入中間的區間時，消費者會減少其消費支出。[[abstract]]In recent years, gold markets have attracted considerable attention from investors, because it is used as hedge to fight the increasing risk after the financial crisis of 2008. The exchange rate of the dollar/euro is interchangeably used in investment portfolios that include precious metals and oil. Furthermore, Crude oil is the one of the most important energy resources in the world and influences the economy and financial markets. In the first essay, we employ smooth transition autoregressive (STAR) model to investigate the nonlinear dynamic behavior of gold price, oil price, and the dollar/euro. The empirical results indicate that these variables exist in a nonlinear relationship. We consider the change rate of the dollar/euro as the threshold value that represents the dynamic LSTAR-type process, whereas the oil price as the threshold value represents the smooth symmetric ESTAR-type process. The results present the greater threshold effect that exists between the gold price and the change rate of the dollar/euro. In the second and third assay, we apply the PSTR model to analysis threshold effect of oil price changes and private consumption expenditures between oil-exporting and oil-importing countries. We use the oil price of one lag period as the threshold variable to explore the nonlinear relationship. The results indicate the value of threshold variable in the low and high regimes will cause consumer to decrease consumption expenditures in oil-exporting countries. When we consider the oil-importing countries, oil price changes will induce the consumption expenditure increase insensitively in the low and high regimes. On the contrary, the oil price changes will reduce the consumption expenditures in the middle regime.[[tableofcontents]]Contents Chapter 1 Introduction 1 Chapter 2 Literature Review 4 2.1 Relationship among gold price, crude oil price, and the exchange rate 4 2.2 Channels of oil transmission - theory 5 2.3 Relationship of oil price and personal consumption in oil - exporting 7 Chapter 3 Methodology 9 3.1 Unit Root Test 9 3.1.1 ADF Test 9 3.1.2 KPSS Test 10 3.2 Smooth Transition Autoregressive (STAR) 11 3.2.1 Building the Smooth Transition Autoregressive (STAR) 11 3.2.2 Testing Linearity 12 3.2.3 Choosing between ESTAR and LSTAR model 13 3.3 Panel Unit Root Test 13 3.3.1 Levin, Lin and Chu (LLC) (2002) Panel Unit Root Test 13 3.3.2 Im, Pesaran and Shin (IPS) (2003) Panel Unit Root Test 14 3.4 Panel Smooth Transition Regression (PSTR) model 14 3.5 Building panel smooth transition regression models 16 3.5.1 Estimation of parameters 18 3.5.2 Evaluation of the estimated model 19 Chapter 4 Empirical Results 20 4.1 Relationship of Threshold Effect among Gold, Oil, and Exchange Rate 20 4.1.1 Data and Summary Statistics 20 4.1.2 Empirical results 20 4.2 Threshold relationship between oil prices and personal consumption expenditures : the case of oil – exporting countries 26 4.2.1 Data and Summary Statistics 26 4.2.2 Empirical results 27 4.3 Threshold relationship between oil prices and personal consumption expenditures : the case of oil - importing countries 32 4.3.1 Data and Summary Statistics 32 4.3.2 Empirical results 33 Chapter 5 Conclusions…………………………………………………37 References 40 List of Tables Table 4.1.1 Unit Root Tests 21 Table 4.1.2 The Results of Delay Length and Linearity Tests 22 Table 4.1.3 The Model Specification for the ESTAR and the LSTAR Model 22 Table 4.2.1 Panel Unit Root Tests 27 Table 4.2.2 Wald (LM) tests for Nonlinearity 28 Table 4.2.3 Determination of the Number of Location Parameters 28 Table 4.2.4 Parameter Estimates for the PSTR Models 30 Table 4.2.5 Relationship between the consumption and oil prices 31 Table 4.3.1 Panel Unit Root Tests 33 Table 4.3.2 Wald (LM) tests for Nonlinearity 34 Table 4.3.3 Determination of the Number of Location Parameters 34 Table 4.3.4 Parameter Estimates for the PSTR Models 36 Table 4.3.5 Relationship between the consumption and oil prices 36 List of Figures Fgure 4.1.1 Transition Function of the LSTAR Model with respect to Transition Variable for the Exchange Rate 23 Figure 4.1.2 Transition Function of the ESTAR Model with respect to Transition Variable for the Oil Price 25[[note]]學號: 897530027, 學年度: 10