本文旨在探討控制設計，而誤差項採用自我迴歸移動平均隨幾模式，損失函數分為平 方誤差與一般函數兩種。全文一冊，共分五章，約三萬餘子。內容如下： 第一章 導論：說明控制設計之目的，理想控制設計的條件，及本文的結構。 第二章 自我迴歸移動平均隨機模式：說明模式的理論基礎，性質，應用及模式的建 立。 第三章 動態系統隨機模式：說明模式的性質，建立，及應用。 第四章 控制設計：分為前饋控制、回饋控制，及一般損失函數的控制。 第五章 結論：說明本文所採用方法的利弊。

本論文包括3篇時間序列方法在財務的應用。第一篇以門檻向量自我迴歸模型(threshold vector autoregression)分析股市訊息傳遞的不對稱效果；第二篇利用不對稱共整合模型(asymmetric cointegration)分析中國大陸股市之間長期均衡關係；第三篇根據追蹤資料共整合檢定(panel Cointegration test)檢定購買力平價說。 第一篇文章利用門檻向量自我迴歸模型分析Nasdaq股市對台灣、日本與韓國股市不對稱的訊息傳遞效果。實證結果發現，當Nasdaq市場處於下跌狀態時（壞消息狀態），Nasdaq市場干擾對亞洲股市的衝擊較大，反之，當Nasdaq市場處於上漲狀態時（好消息狀態）時，Nasdaq市場干擾對亞洲股市的衝擊較小，而在壞消息狀態時，Nasdaq指數大跌對Jasdaq指數與Kosdaq指數的衝擊效果大於Nasdaq指數大漲的效果，Nasdaq指數小跌所產生的衝擊與小漲所產生的效果具有對稱性。 第二篇文章以Enders and Siklos(2001)不對稱共整合模型探討，中國大陸上海及深圳A股與B股股價指數之間長期不對稱的均衡關係，實證結果發現，在1992年10月至2001年8月，上海A股指數與深圳A股指數之間具有不對稱共整合關係，且當上海A股處於好消息狀態（股市上漲）時，其誤差修正項的調整速度較壞消息狀態（股市下跌）之下為快，此外，上海A股指數與深圳A股指數之間其有雙向的連動關係。在B股開放之後，則是深圳股市A股與B股指數存在不對稱共整合關係，同時Granger因果關係檢定顯示深圳B股指數領先A股指數。 第三篇文章利用Pedroni(2001)追蹤資料共整合檢定，探討大麥克漢堡價格與CPI兩種不同的價格指數用於檢定購買力平價說的有效性，根據14個國家1992-1999年的追蹤資料得到的實證結果顯示，以名目匯率作為被解釋變數，則大麥克漢堡價格與CPI都是支持PPP假說，然而若以相對價格為被解釋變數，則只有大麥克漢堡價格是支持PPP假說，而以CPI為基礎的PPP假說則是無法得到支持。除此之外，本文的實證結論並不受生產力差異的影響。 關鍵字：門檻向量自我迴歸模型、不對稱共整合、追蹤資料共整合、股票市場、購買力平價說 / This dissertation includes three financial applications using time series methods. The first article investigates the asymmetric effects of information transmissions in stock markets using threshold vector autoregression model. The second article uses asymmetric cointegration to study the long-run equilibium relationships among Chinese stock markets. The third article uses panal cointegration to test purchasing-power parity (PPP). Firstly, we examines the asymmetric effects of information transmissions of Nasdaq stock market on Taiwan, Japan, and Korea stock markets by using a threshold vector autoregressive model. And also, we check whether Nasdaq stock market have different impacts on organized stock exchanges (including TAIEX, NIKKEI 225 Index, Korea Composite Index) and over-the-counter markets (including Taisdaq Index, Jasdaq Index, and Kosdaq Index) or not. The empirical results indicate that negative innovations in Nasdaq market (bad news regime) have large influence on Asia stock markets. Particularly, the positive innovations in Nasdaq market (good news regime) have small influence on Asia stock market. The large negative innovations in Nasdaq market have great influence than those of the large positive innovations on Jasdaq Index and Kosdaq Index in bad news regime. The second article uses Enders and Sikios's (2001) asymmetric cointegration model to investigate the long-run asymmetric equihbrium relationships. The empirical results find that there exits an asymmetric cointegrated relationship between Shanghai A share index and Shenzhen A share index for the period from October 1992 to August 2001. The adjustment parameters of error correction term at Shanghai A share market are larger in bad-news regime than those in good-news regime. This result reveals investors at Shanghai possess over-reaction behavior on news of stock market. Moreover, there exists a bi-directional Granger causality between Shanghai A share index and Shenzhen A share index. We find there exists an asymmetric cointegrated relationship between Shenzhen A share index and Shenzhen B share index after 19 February 2001. Furthermore, the Shenzhen B share index leads Shenzhen A share index after 19 February 2001. The third article uses Pedroni's (2001) panel cointegration test to examine the validity of PPP hypothesis by two different price indces, i.e. Big Mac prices and CPI. Our panel observations include 14 countries from 1992 to 1999. The empirical evidence indicates Big Mac PPP and CPI PPP is supposed if we use nominal exchange rate as the explanatory variable. Nevertheless, the Big Mac PPP is valid but CPI PPP not valid if we use price level as the explanatory variable. Moveover, our concludtion does not influenced by productivity bias. Keywords: threshold vector autoregression, asymmetric cointegration, panel cointegration, stock markets, purchasing-power parity

本研究之宗旨在於探究中國大陸對亞洲區域內國家的金融市場影響程度之變化。由過去的各國股市日報酬率資料間相關程度與政策改變間的影響結果，來觀察是否未來在兩岸政策更開放下會使中國大陸對台灣的影響程度上升，進而使國際間投資組合的風險分散效果下降。本研究自DataStream選取台灣、香港、中國大陸、泰國、印尼、新加坡、馬來西亞、菲律賓、日本以及美國等十國的股價指數日資料，以對數轉換為日報酬率後年化加以分析。選取時間自1991年7月15日(中國大陸上海證券交易所股價指數公開後)至2008年12月31日。本研究選用的方法為使用風險值(VaR; Value at Risk)的概念來取代傳統的標準差，衡量以該十國所分別組成的各投資組合風險值變動情形；以及由風險值所衍生出的Diversification Benefit與Incremental VaR的結果。發現到僅由亞洲區域國家內組成的投資組合風險分散效果逐漸下降；且效果並不如有納入區域外國家(如美國)的投資組合。接著本研究將Gaussian Copula模型放入VaR中以增加對極端值的捕捉能力，結果發現本研究所選用的指數加權移動平均法所求得之相關係數已可有效反應出各國之間的相依程度，即加入Copula的效果有限。另外藉由Copula所求得之相關係數顯示，台灣、香港對中國大陸之間的相依程度已逐漸上升，並開始出現超越美國之現象，其中又以2005年為上升趨勢的起點。最後本研究以向量自我迴歸模型(VARs)來驗證2005年前後中國大陸股市對其他亞洲區域國家的影響力是否存在結構性的改變；並再佐以變異數拆解之方法來觀察2005年前後各國家之間自發性衝擊對彼此之間的影響程度變化。研究結果發現，透過VARs可證明中國大陸對亞洲區域各國的影響力在2005年後轉變為顯著；僅對美國不存在此一現象。另外變異數拆解的結果也顯示各國之間的相依程度在2005年後有明顯的上升，中國大陸對各國的影響程度亦然。透過本研究之結論，在未來兩岸將簽訂金融監理備忘錄使整合關係提升的環境下，需提醒投資人整合關係的上升將使得以之為標的之投資組合風險分散效果下降，需作為投資策略之考量。 / The object of this research is to find out the trend of dependence and correlation between China and other Asian countries. Based on past information about the relationship between equity markets’ correlation and changes in policies, this research can make suggestions to the foreseeable future of Taiwan and China whose relationship will be more solid due to new policy. The data of this research are gathered from DataStream, which includes Taiwan, Hong Kong, China, Thailand, Indonesia, Singapore, Malaysia, Philippines, Japan and United States. Selected from 1991/07/15 (when the Shanghai SE Composite went public) to 2008/12/31, this research calculates the annualized daily return using natural logarithms of two consecutive daily index prices. This research uses Value at Risk (VaR) to measure the risk exposure of portfolios formed by ten countries, and extends to the use of Diversification Benefit and Incremental VaR. The results found out that the diversification effects of portfolio which includes only Asian countries are decreasing and inferior to the effects when cross region countries are included. The second study of this research is to combine Gaussian Copula Model with VaR to capture the effects of extreme values. Empirical results found out that the VaR using Exponentially Weighted Moving Average method is good enough for analyzing Asian stock markets. The correlation in Copula model suggests that the dependence between Taiwan and China had increased since 2005 and has the increasing trend which might overwhelm the dependence between Taiwan and United States. Final research is about using Vector Autoregressions Model (VARs) to testify is there exist any structural change of dependence before and after 2005, and using Variance Decomposition to observe the relationships between these ten countries. The results found out that there exist structural change in 2005, the post-2005 periods shows that for Asian countries the effect from China are significant and greater than pre-2005 periods.

本論文的研究主軸圍繞於自我迴歸模型的估計與推論上。文獻上自我迴歸模型的估計多直接採用最小平方法, 但此估計方式卻有兩個缺點:(一)當序列具單根時,最小平方估計式的漸近分配為非正規型態,因此檢定時需透過電腦模擬得到臨界值;(二)最小平方估計式雖具一致性,但卻有嚴重的有限樣本偏誤問題。有鑑於此,我們提出一種「二階差分轉換估計式」,並證明該估計式的偏誤遠低於前述最小平方估計式,且在序列為粧定與具單根的環境下具有相同的漸近常態分配。此外,二階差分轉換估計式相當適合應用於固定效果追蹤資料模型,而據以形成的追蹤資料單根檢定在序列較短的情況下仍有不錯的檢定力。 本論文共分四章,茲分別簡單說明如下: 第1章為緒論,回顧文獻上估計與推論自我回歸模型時的問題,並說明本論文的研究目標。估計自我迴歸模型的傳統方式是直接採取最小平方法,但在序列具單根的情況下由於訊息不隨時間消逝而快速累積,使估計式的收斂速度高於序列為恒定的情況。不過,這也導致最小平方估計式的漸近分配為非標準型態,並使得進行假設檢定前必須先透過電腦模擬來獲得臨界值。其次,最小平方估計式雖具一致性,但在有限樣本下卻是偏誤的。實證上, 樣本點不多是研究者時常面臨的窘境,並使得小樣本偏誤程度格外嚴重。本章中透過對前述問題形成因素的瞭解,說明解決與改善的方法,亦即我們提出的「二階差分轉換估計式」。 第2章主要目的在於推導二階差分轉換估計式之有限樣本偏誤。我們亦推導了多階差分自我迴歸模型下二階段最小平方估計式(two stage least squares, 2SLS)與 Phillips andHan (2008)採用的一階差分轉換估計式之偏誤,以同時進行比較。本章理論與模擬結果皆顯示,一階與二階差分轉換估許式與2SLS之 $T^{−1}$ 階偏誤程度皆低於以最小平方法估計原始準模型(level model)的偏誤,其中 T 為時間序列長度。另外,一階差分轉換估計式與二階差分轉換估計式在 $T^{−1}$ 階偏誤上,分別與一階和二階差分模型下2SLS相同,但兩估計式的相對偏誤程度則因自我相關係數的大小而互有優劣。同時,我們發現估計高於二階的差分模型對小樣本偏誤並無法有更進一步的改善。最後,即使在樣本點不多的情況下,本章所推導的偏誤理論對於實際偏誤仍有良好的近似能力。 第3章主要目的在於發展二階差分轉換估計式之漸近理論。與 Phillips and Han (2008) 採用之一階差分轉換估計式相似的是,該估計式在序列為恒定與具單根的情況下收斂速度相同,並有漸近常態分配的優點。值得注意的是, 二階差分轉換估計式的漸近分配為 N(0,2),不受任何未知參數的影響。另外,當序列呈現正自我相關時,二階差分轉換估計式相較於一階差分轉換估計式具有較小的漸近變異數,進而使得據以形成的檢定統計量有較佳的對立假設偵測能力。最後, 誠如 Phillips and Han (2008) 所述,由於差分過程消除了模型中的截距項,使得此類估計方法在固定效果的動態追蹤資料模型(dynamic panel data model with fixed effect) 具相當的發展與應用價值。 本論文第4 章進一步將二階差分轉換估計式推展至固定效果的動態追蹤資料模型。文獻上估計此種模型通常利用差分來消除固定效果後,再以一般動差法 (generalized method of moments, GMM) 進行估計。然而,這樣的估計方式在序列為近單根或具單根時卻面臨了弱工具變數(weak instrument)的問題,並導致嚴重的估計偏誤。相反的,差分轉換估計式所利用的動差條件在近單根與單根的情況下仍然穩固,因此在小樣本下的估計偏誤相當輕微(甚至無偏誤)。另外,我們證明了不論序列長度(T )或橫斷面規模(n)趨近無窮大,差分轉換估計式皆有漸近常態分配的性質。與單一序列時相同的是,我們提出的二階差分轉換估計式在序列具正自我相關性時的漸近變異數較一階差分轉換估計式小;受惠於此,利用二階差分轉換估計式所建構的檢定具有較佳的檢力。值得注意的是,由於二階差分轉換估計式在單根的情況下仍有漸近常態分配的性質,我們得以直接利用該漸近理論建構追蹤資料單根檢定。電腦模擬結果發現,在小 T 大 n 的情況下,其檢力優於文獻上常用的 IPS 檢定(Im et al., 1997, 2003)。 / This thesis deals with estimation and inference in autoregressive models. Conventionally, the autoregressive models estimated by the least squares (LS) procedure may be subject to two shortcomings. First, the asymptotic distribution of the LS estimates for autoregressive coefficient is discontinuous at unity. Test statistics based on the LS estimates thus follow nonstandard distributions, and the critical values obtained need to rely on Monte Carlo techniques. Secondly, as is well known, the LS estimates of autoregressive models are biased in finite samples. This bias could be substantial and leads to serious size distortion for the test statistics built on the estimates and forecast errors. In this thesis,we consider a simple newmethod ofmoments estimator, termed the “transformed second-difference” (hereafter TSD) estimator, that is without the aforementioned problems, and has many useful applications. Notably, when applied to dynamic panel models, the associated panel unit root tests shares a great power advantage over the existing ones, for the cases with very short time span. The thesis consists of 4 chapters, which are briefly described as follows. 1. Introduction: Overview and Purpose This chapter first reviews the literature and states the purpose of this dissertation. We discuss the sources of problems in estimating autoregressive models with the conventional method. The motivation to estimate the autoregressive series with multiple-difference models, instead of the conventional level model, is provided. We then propose a new estimator, the TSD estimator, which can avoid (fully or partly) the drawbacks of the LS method, and highlight its finite-sample and asymptotic properties. 2. The Bias of 2SLSs and transformed difference estimators in Multiple-Difference AR(1) Models In this chapter, we derive approximate bias for the TSD estimator. For comparisons, the corresponding bias of the two stage least squares estimators (2SLS) in multiple-difference AR(1) models and the transformed first-difference (TFD) estimator proposed by Chowdhurry (1987) are also given as by-products. We find that: (i) All the estimators considered are much less biased than the LS ones with the level regression; (ii)The difference method can be exploited to reduce the bias only up to the order of difference 2; and (iii) The bias of the TFD and TSD estimators share the same order at $O(T^{-1})$ as that of 2SLSs. However, to the extent of bias reductions, neither the 2 considered transformed difference estimators shows a uniform dominance over the entire parameter space. Our simulation evidence lends credible supports to our bias approximation theory. 3. Gaussian Inference in AR(1) Time Series with or without a Unit Root The goal of the chapter is to develop an asymptotic theory of the TSD estimator. Similar to that of the TFD estimator shown by Phillips and Han (2008), the TSDestimator is found to have Gaussian asymptotics for all values of ρ ∈ (−1, 1] with $\sqrt{T}$ rate of convergence, where ρ is the autoregressive coefficient of interest and T is the time span. Specifically, the limit distribution of the TSD estimator is N(0,2) for all possible values of ρ. In addition, the asymptotic variance of the TSD estimator is smaller than that of the TFD estimator for the cases with ρ > 0, and the corresponding t -test thus exhibits superior power to the TFD-based one. 4. Estimation and Inference with Moment Methods for Dynamic Panels with Fixed Effects This chapter demonstrates the usefulness of the TSD estimator when applying to to dynamic panel datamodels. We find again that the TSD estimator displays a standard Gaussian limit, with a convergence rate of $\sqrt{nT}$ for all values of ρ, including unity, irrespective of how n or T approaches infinity. Particularly, the TSD estimator makes use of moment conditions that are strong for all values of ρ, and therefore can completely avoid the weak instrument problem for ρ in the vicinity of unity, and has virtually no finite sample bias. As in the time series case, the asymptotic variance of the TSD estimator is smaller than that of the TFD estimator of Han and Phillips (2009) when ρ > 0 and T > 3, and the corresponding t -ratio test is thus more capable of unveiling the true data generating process. Furthermore, the asymptotic theory can be applied directly to panel unit root test. Our simulation results reveal that the TSD-based unit root test is more powerful than the widely used IPS test (Im et al, 1997, 2003) when n is large and T is small.

本文主要在探討影響台灣股票市場波動的因素，除了考慮以之前學者設定的 VAR(12)模型研究，另外以 SUR(5)模型來討論股市波動與基本面、交易面間的關係；最後，再以自我迴歸異質條件變異數模型來分析股市波動的特性。最重要的是，我們會根據誤差項的各類檢定結果來判定研究股市波動性質的最佳模型。 在聯立方程式的估計中，我們發現代表資訊到達指標的兩變數--週轉率與成交量成長率--會影響股票市場的波動。另外，我們找出交易面（成交量成長率）可能會影響基本面（匯率），這也就是說，在研究股市波動時，我們不需要特別區分變數的屬性。 在 GARCH 模型及 TGARCH 模型中，我們仍然可發現週轉率與成交量成長率會影響股市條件平均數或條件變異數；除此之外，好壞消息對股市日報酬率條件變異數（條件波動）應有不同的影響效果（壞消息的影響力較快反應）。而股市自身風險係數雖然統計檢定上不顯著異於零，但若未加入條件平均數的估計式，則可能會使模型得到較差的誤差項檢定結果，顯見股市自身風險應為影響投資人設定期望報酬率水準的重要因素之一。 從上述估計結果，我們可以知道，若散戶投資人能正確解讀市場上出現的各種新資訊之背後意義，將可使成交量成長率或週轉率（大部份可能代表無意義或不正確的交易行為）的變動幅度降低，進而有效地減少股票市場中股價異常波動的現象。 / My essay's topic focuses on discussing the factors that influence stock market volatility in Taiwan's stock market. Besides VAR(12) model as previous researchers have studied, I tries to set up SUR(5) models analyzing the relationship among the stock market volatility、the foundamental variables'volatilities and trading activities; Then I cited ARCH models ( autoregressive conditional heteroskedisticity models ) to find out the characteristics of stock market volatility. Most important of all, according to each misspecification test ( residual test ), I would specify the better models to describe the stock market volatility. In the estimations of system equations ( VAR(12)and SUR(5)models ), first I found that turnover rate and the growth rate of trading volume, which represent the information arrival indexes, could effect stock return's monthly conditional variance. Second, I especially found out the evidence that trading activities (trading volume growth) would probably have an impact on the macroeconomic variable ( exchange rate volatility ). It shows that we don't need to distinguish the attributes of those factors which could influence stock market volatility. In GARCH and TGARCH model, the positive influences of turnover and trading volume growth on daily stock return's conditional mean and conditional variance ( conditional volatility ) are still obvious, Within these TGARCH model, I discovered that bad news and good news could have different influences on stock market volatility ( the impact of bad news which resulted in downward movements of stock market volatility appeared faster that the good news'which caused upward movements). Stock market's self-risk(σ_t-1^{^2}) is statistically insignificant different from zero in GARCH models, but when I omitted this variable in daily stock return's conditional mean estimation equation, standardized residual might not obey the assumption of normal distribution. It apparently told us that the stock market's self-risk term ( σ_t-1^{^2} ) is one of the critical factors which influences investors to estimate expected return level. From those results above, we realized that if investors could precisely understand the real meanings of new information conveying in the stock market, it might decrease the levels of turnover and trading volume growth ( which could sometimes represent meaningless or inexact trading activities ), then effectively reduce the abnormal volatility phenomenon in stock market.

犯罪行為受到警力或法律執行的影響，會移動到鄰近警力較差地區。正如 Gylys所說：考量一個地區警力的多寡，將受到其他鄰近區域警力的影響 很大。Mehay亦認為：從實際經驗上來看，對於移動性的犯罪(如搶劫、縱 火、偷竊等)，外部支配型式力量(如警力)的適當增加，將迫使其外溢( spillovers)至鄰近區域。利用空間統計的自我迴歸模式，我們可以更了 解移動性犯罪受到相連區域自我相關的影響。即使相關性不高，在作了差 分之後，其主成分分析最大負載變數項，變化相當大。所以資料裡，如果 有區域自我相關的情形，就應該謹慎處裡。

本研究的目的在探討台灣貨幣市場短天期利率衍生性金融商品（30天期商業本票利率期貨、90天遠期利率協定〈以下簡稱FRA〉）對其即期利率（30天期商業本票利率、90天期商業本票利率）之『價格發現』功能。可由兩方面來檢定利率遠期協定或利率期貨市場之『價格發現』功能：（1）市場效率性：FRA、利率期貨價格可否作為未來到期日時即期利率之不偏預期；（2）FRA、利率期貨與即期利率價格間之領先-落後關係。 選取各交易日的日資料作為觀察值。在研究方法上採用ADF單根檢定、效率性檢定、向量自我相關模型（VAR）、Granger因果關係檢定、誤差正交檢定、共整合檢定、誤差修正模型。 結論結果發現，90天遠期利率協定（FRA）對90天期商業本票利率進行『價格發現』的分析，以「市場效率性檢定」的結果顯示此市場無效率，亦即無『價格發現』，可能是因為買FRA的機構投資人目的不是持有到到期，僅為判斷短期利率走勢方向，可能買個幾天欲賺取差價利潤，所以非為未來現貨價格的不偏預期；以「領先-落後關係分析」，顯示其無『價格發現』，此一結果的可能解釋是由於台灣FRA市場非集中市場公開交易，交易量尚不及現貨市場。因此市場資訊的不透明可能使遠期契約價格不如現貨價格般具代表性。 30天期商業本票利率期貨對30天期商業本票利率進行『價格發現』的分析，以「市場效率性檢定」結果顯示在到期日前適當的期間（24~36天）此市場具有效率性，即存在『價格發現』；而「領先-落後關係分析」結果則無明顯的領先落後，不具有期貨領先現貨的『價格發現』，此部分我們可能提出的解釋為：在30天期商業本票利率期貨剛推出不久，一般市場上的交易者大多是從事避險交易，鮮少進行投機行為，所以不具有短天的領先落後關係，其顯示價格發現是在考慮市場存在風險溢酬下，到期前24~36天的利率期貨價格是未來現貨價格的預期。

本論文主要是探討在二維度空間上二母體分佈是否一致。我們利用排列 (permutation)檢定方法來做比較, 並藉由費雪(Fisher)正確檢定方法的想法而提出重標記 (relabel)排列檢定方法或稱為費雪排列檢定法。 我們透過可交換性的特質證明它是正確 (exact) 的並且比 Syrjala (1996)所建議的排列檢定方法有更高的檢定力 (power)。 本論文另提出二個空間模型: spatial multinomial-relative-log-normal 模型 與 spatial Poisson-relative-log-normal 模型 來配適一般在漁業中常有的右斜長尾次數分佈並包含很多0 的空間資料。另外一般物種可能因天性或自然環境因素像食物、溫度等影響而有群聚行為發生, 這二個模型亦可描述出空間資料的群聚現象以做適當的推論。 / This thesis proposes the relabel (Fisher's) permutation test inspired by Fisher's exact test to compare between distributions of two (fishery) data sets locating on a two-dimensional lattice. We show that the permutation test given by Syrjala (1996} is not exact, but our relabel permutation test is exact and, additionally, more powerful. This thesis also studies two spatial models: the spatial multinomial-relative-log-normal model and the spatial Poisson-relative-log-normal model. Both models not only exhibit characteristics of skewness with a long right-hand tail and of high proportion of zero catches which usually appear in fishery data, but also have the ability to describe various types of aggregative behaviors.

台灣不動產投資信託（T-REITs）自2005年發行至今已逾六年，然其市場表現仍不如發行之初所預期。過去國內已有許多研究針對T-REITs市場發展進行探討，然而目前就T-REITs與直接不動產投資市場價格表現間之相關研究尚付之闕如。有鑑於此，本研究藉由共整合與Granger因果關係檢定，檢視REITs與直接不動產市場間之關聯性，了解台灣與美國之REITs市場表現差異及其影響因素，進而作為改進T-REITs運作機制或架構之參考依據。 實證結果發現，美國之REITs與直接不動產市場之間存在共整合關係。此結果表示，長期而言，這兩者可能具有相似之風險分散效益。此外，透過Granger因果關係檢定發現REITs領先於直接不動產，乃因前者市場較具效率。另一方面，台灣之REITs與直接不動產市場之間則不具有共整合以及領先或落後關係，然直接不動產當期價格仍會受到本身與REITs之前期價格影響。 本研究進一步分析台、美兩國實證結果之差異原因如下：資料的樣本期間、REITs市場規模、存在於T-REITs市場之集中性風險以及潛在的代理問題。其中，針對T-REITs潛在代理問題，本研究藉由分析股票與T-REIT報酬率之波動性，發現T-REIT之不動產管理機構若與母集團相關者，則其市場表現較差。因此，我們得出T-REITs市場發展主要是受限於代理問題之結論。本研究成果不僅有助於改善T-REITs市場效率，亦可提供學術與實務之參考。 / The mechanism of Real Estate Investment Trusts in Taiwan (or T-REITs) was launched in 2005, however, T-REITs market did not perform as expected. What caused the limited development of T-REITs market? Current literature on the performance between T-REITs and direct real estate investment is limited. Through the cointegration and Granger causality tests, the purpose of this study is hence to explore the short-term and long-term dynamics between REITs and direct real estate markets in the U.S. and Taiwan, respectively. This study presents evidence of the cointegration relationship between REITs and direct real estate in the U.S. It implies that the diversification properties of these two assets are likely to be similar over the long horizon. According to the Granger causality test, REITs leads direct real estate due to the market information efficiency. These findings are consistent with those of previous studies. On the other hand, we find no cointegration and lead-lag relation between T-REITs and commercial real estate. Moreover, the current commercial transaction price is affected by both its and T-REIT previous price. By comparing the difference between the results of these two countries, there are several possible explanations for the different results between the U.S. and Taiwan, including difference in sample period, market capitalization, concentrated risk, and most importantly, the potential agency problem existing in T-REITs market. Finally, the underperformance of parent-related management T-REIT is verified through the volatilities of stock and T-REIT returns. Therefore, we conclude that the limited development of T-REITs is caused by the agency problem in REITs market. Results of this study may provide T-REITs market for improving its efficiency, as well as for the reference for both academics and real practices.