本文以貝氏方法估計方向距離函數，加入單調性與曲度限制，同時考慮環境變數於模型中。為了凸顯考慮非意欲產出方向距離函數的優點，本文同時估計產出面射線距離函數，並與方向距離函數模型比較。實證分析資料為1970至2010年間各國總體經濟變數，分別在有無加入限制條件與環境變數的狀況下，估計兩種距離函數，從無效率值、效率分數與技術進步率等角度分析彼此間的差異。發現射線距離函數模型由於忽略非意欲產出，傾向高估生產單位的技術效率。另一方面，其係數估計值容易違反射線距離函數的先天性質，較不具參考性。而方向距離函數模型當中，有無加入限制條件與有無考慮環境變數的模型結果各不相同，其中同時加入限制條件與環境變數的模型結果最為合理。

本研究使用多種資料蒐集方法，對全球經濟大國中國大陸海外高層次人才引進計畫（簡稱千人計畫）之數據資料進行量化分析，本研究針對千人計畫第一批至第五批部分專家學者，分析在2009年至2014年間透過此計畫資源投入的整體學術研究產出效率，並進一步討論不同研究機構在資源獲得及學術產出效率上的差異。模型運算結果後發現，首先，國家自然科學基金在千人計畫研究經費配置上明顯偏向國家研究機構，其次，國家研究機構不論在研究產出效率或生產力指標上表現都較中國大學佳，這樣的結果說明了，千人計畫資源錯置的情況確實存在，若集中研究資源並合理配置給國家研究機構進行研發創新，將資源分配給產值高、研究成果豐碩的學者們，這樣對於中國科學領域巨額投入的回報只會有增無減。本文至此，可對中國菁英階層主導的千人計畫政策，就學術產出效率層面，建議最適的人力資本投資方向，以期能避免嚴重的資源錯置問題，並且對後續相關研究者有所助益。 / This research measures the resource misallocation of human capital development in China. We create a dataset composed of Chinese repatriate researchers of the first to fifth groups of Thousand Talent program scientists during 2009 - 2014. The result shows that National Research Institutions have long outperformed university in productivity and research output efficiency. This research; thus, contribute to evaluate the impact brought about by China's overseas high-level talents’ introduction policy via quantitative analyses and propose a variety of data collection techniques to create a reliable dataset for future research.

創新是複雜、昂貴、且高風險的活動，並且存在外部性，研發合作為促使企業從事創新的重要機制。本研究目的在於延伸過去理論性架構與實證研究，建立研發合作—創新—財務績效價值鏈。以下為研究問題： 1.吸收能力、知識外溢、與不確定性是否會影響研發合作的頻率？ 2.研發合作是否可以提高研發投資、研發產出、與財務績效？ 3.不同的研發合作型態如何影響研發合作的決定因素？ 4.不同的研發合作型態如何影響研發合作的績效？ 5.研發合作與財務績效的關係是否會受到研發投資與研發產出的中介影響？ 本研究採用 two-industry, n-firm-per-industry Cournot 競爭模型探討研發合作、研發投資（創新之投入面）、研發產出（創新之產出面—非財務績效）、與財務績效的關係，並以台灣高科技產業為研究對象進行實證分析。對於台灣高科技產業的研發合作與創新活動，研究結果提供學術界與企業界更完整且廣泛的觀點。 實證結果支持公司擁有較高吸收能力的員工是從事研發合作的決定因素之一。另外，知識外溢的提高，亦將促使高科技公司進行研發合作。而在高度吸收能力與知識外溢下，公司採行一般合作之頻率較其他合作模式高。 另外，實證結果也發現研發合作的確鼓勵台灣高科技產業的公司進行更多研發的投資，並且持續創造較高的研發產出與財務績效。相對於其他合作型態，一般合作可以創造較高的研發產出與財務績效，因此為較佳的合作模式。而由於市場競爭的本質，使得水平合作公司之研發投資較垂直合作與一般合作少。最後，僅有研發投資並不足以提升公司的績效與維持競爭優勢，研發合作公司的創新能力與研發產出才是獲利力的決定因素。 / Innovation is complex, costly, and risky and incurs externalities. R&D cooperation is thus a proper mechanism to encourage firms to innovate. The purposes of this dissertation are to extend the prior theoretical framework and empirical studies to establish a research framework for the R&D cooperation—innovation—financial performance chain. The research questions are as follows: 1.Do absorptive capacity, knowledge spillovers, and uncertainty affect the intensity of R&D cooperation? 2.Does R&D cooperation result in higher R&D investments, R&D outputs, and financial performance? 3.How do different R&D cooperation types influence the determinants of R&D cooperation? 4.How do different R&D cooperation types influence the performance of R&D cooperation? 5.Is the effect of R&D cooperation on financial performance mediated by R&D investments and R&D outputs? In this dissertation I apply the two-industry, n-firm-per-industry Cournot competition models to theoretically examine the relationship between R&D cooperation, R&D investments (input perspective of innovation), R&D outputs (output perspective of innovation—non-financial performance), and financial performance. I then use Taiwan’s high-technology industry as a research sample and empirically test my research hypotheses. The results provide academia and practitioners with a more comprehensive view of R&D cooperation and innovation activity among Taiwan’s high-technology industries. The empirical results support the argument that absorptive capacity has a positive impact on the frequency of R&D cooperation in high-technology industry. In addition, an increase in knowledge spillovers also tends to increase intensity to collaborate in R&D. Under high absorptive capacity and knowledge spillover, generalized R&D cooperation is preferred to other cooperative models. The empirical results also show that R&D cooperation does encourage Taiwan’s high-technology firms to invest more resources in R&D, and leads to higher R&D outputs and financial performance under the characteristic of high knowledge spillovers. Relative to other cooperation types, generalized cooperation leads to higher R&D outputs and financial performance and is a superior cooperative model. Due to the nature of market competition, horizontal cooperative firms are not willing to invest too much in R&D relative to vertical cooperation and generalized cooperation. Finally, simply investing in R&D alone is not enough to achieve breakthrough performance and sustain a competitive advantage. The ability to innovate and generate R&D outputs determines the profitability of the cooperative company.

海外華人人口，截至2004年底止，約為3,808萬人（不含香港、澳門），政府為照顧為眾多之海外僑胞，先於1926年於國民政府下設僑務委員會以推進僑務，復為擴大服務僑社並增進僑民福址，於1985年針對海外僑社需求，選擇華僑眾多之地區，設置華僑文教服務中心，自1985年起迄2004年底止，僑委會在海外各地前後共設置了十七個文教服務中心。為維持該等中心之運作，政府每年均編有巨額預算挹注，為瞭解相對於投入之資源，中心之產出是否符合經濟學上『投入』與『產出』之生產效率概念，本文之研究爰以產出導向之資料包絡分析法建立實證模型進行探討，同時以各文教服務中心為一決策單位，將2001年至2004年間各文教中心之投入及產出項目投入實證模型中，以計算各文教中心的經營績效。 實證結果顯示，海外文教中心整體的技術效率平均值介於2.4662與3.1605之間，同時規模效率平均值大於1，顯示造成各文教中心技術無效率的來源，大部份來自於純技術效率，另一部分則來自於規模無效率。 同時依實證結果資料可以發現，（一）僑委會截至2004年止之十七個文教中心，在各種不同的組合下，僅有4至9個中心之效率值為1，占全部文教中心的二成至五成，顯示整體經營績效有待加強與提升；（二）以僑委會關閉倫敦及墨爾本文教中心雖與實證分析結果相符，然績效較墨爾本中心為低之雪黎文教中心卻未關閉之情況而言，益顯數據性資料於決策過程中之重要性；（三）以全球華人人口分配來看，北美洲地區文教中心之技術效率值，無論在固定或變動規模下，就各種組合而言，均較其他地區文教中心之效率值為佳，顯示除應提高北美地區以外中心之效率外，似乎也反映了服務海外僑胞之資源分配是否過度集中，及與僑胞互動應予加強之問題。 最後，由於本研究係首次將DEA運用於駐外機構經營效率之評估，因此本文之相關實證模型將可作為評估我國其他部會所屬駐外機構效率衡量之基礎及參考。 / The overseas Chinese population, up till 2004, approximates 3.8 millions (not counting Hong Kong, Macao). To attend to their needs, the government has first established the Commission of Overseas Chinese Affairs under the Nationalist Government in 1926. To better and further serve this overseas Chinese community, the government has chosen among several populous areas to set up Chinese Culture and education service center later in 1985. Between 1985 and the end of 2004, 17 service centers have been set up. To keep these centers running requires a big budget support from the government each year. To examine if these centers provide values that conform to the economic concept of "the input" and "the output" production efficiency, this article has used the data envelopment analysis method to establish the real diagnosis model. This model has used various culture and education service centers as policy-making units from 2001 to 2004 to evaluate the efficacy of these service centers. Test results have shown that the mean value of central whole technical efficiency is between 2.4662 and 3.1605. Also, the mean value of scale efficiency is bigger than 1. This demonstrates the inefficient central technology rate of the service centers results mainly from pure technical inefficiency, and partly from diseconomy of scale. On the other hand, test results also show that, (1) up till 2004, among 17 culture and education centers, just only remain 4 to 9 with central efficiency value of 1 under different kinds of combinations. The fact that they account for two tenths to five tenths of the total number of service centers demonstrates the need for improvement in their operating performance (2) the closing of both London and Melbourne culture and education centers does not conform to the findings of the real diagnosis analysis, but Sydney center is not closed, which reveals the importance of data material in the decision-making process; (3) looking at the distribution of global overseas Chinese population, the technology efficiency values for culture and education centers in North America area are far higher than other areas, measured by all sorts of combinations regardless of fixed or variable scale. This has revealed issues of not only efficiency enhancements for local centers excepting this area, but also excessive concentration of service resources allocated within the region and augment the co-activity with compatriots. Finally, since this is the first time DEA has been utilized in a research to evaluate the management efficiencies for government agencies abroad, the real diagnosis model presented in this article might be used as a reference for similar studies in the future.

本文建構代理人基之社會網路新凱因動態隨機一般均衡模型(Social Network-Based DSGE model)，並分別使用權衡性門檻型泰勒法則與一般線型泰勒法則作為代理人基之社會網路新凱因斯動態一般均衡模型中的貨幣政策方程式，模擬產出缺口、通貨膨脹、利率等總體經濟變數資料，接著利用模擬資料，探討不同網路結構對產出缺口、通貨膨脹等總體經濟變數的影響，同時比較權衡性貨幣政策與法則性貨幣政策穩定經濟的有效性。 　　透過產出缺口與通貨膨脹的波動性分析，本研究發現某些特定社會網路結構的影響力大於貨幣政策的影響力，決定了經濟變數的波動程度。在完全連結網路(Fully)的結構下，通貨膨脹與產出缺口的波動度明顯低於其他結構，而無標度網路(Scalefree)的結構會使產出與通膨的波動程度最大。經過驗證，本研究發現群聚度大、平均路徑短的網路結構內節點之間資訊流通速度較快，對穩定經濟有正面助益；相反的，由於無標度網路強大的中心性，使該網路內指標性節點對其餘節點具有龐大影響力，增加節點內決策的不確定性，連帶造成經濟的大幅波動。另外，在相同的網路結構下比較權衡與法則貨幣政策，研究結果指出權衡性政策會造成較大的產出缺口波動，但對抑制通貨膨脹波動的效果較佳；相對的，法則性政策對產出缺口的穩定效果較好，但卻無法兼顧通貨膨脹的波動性。 / We construct an agent-based New Keynesian DSGE model (Dynamic Stochastic General Equilibrium) with different social network structures to investigate the effects of the rule and discretion monetary policy. According to our simulation results, we find the economic stability depends on the specific social network structure rather than the monetary policy basis like rule and discretion. Generally speaking, the more average path length (the less average clustering coefficient) the network structure is, the more economic fluctuation would be. Also, the results show that scalefree network will lead the most dramatic economic fluctuations. These results are ascribed to scale -free’s high centrality. However, if the social network structure is too complicate to control, the central banker can only manipulate the monetary policy to stabilize the economy. With different policy basis, we find the rule monetary policy will lead less output gap volatility.

本文以2003到2006年美國上市之不動產投資信託（REITs）的資產稅賦遞延交換做研究，並用資產出售交易作為比較，觀察稅賦遞延交換對股票報酬和股利的影響，研究結果發現稅賦遞延交換對於股票報酬有負的宣告效果，然而出售資產的交易有正的且顯著的宣告效果，由於美國REITs基於稅法規定，作為免稅體，每年要以股利的形式分配百分之九十的盈餘給股東，稅賦遞延交換並不能像資產出售交易一樣帶來現金流入，因此對於未來股東的股利所得有所影響，股東對於股票報酬沒有正向的反應，但是股東會考慮稅賦遞延交換會帶來資產重配置的效率，再加上REITs通常會支付比規定還要多的股利，因此稅賦遞延交換的對於股票報酬的負影響會因此而減弱，進一步針對交易方式還有REITs股利分配進行研究，研究的結果支持稅賦遞延交換後的股利比起直接出售交易後所發放的股利還要少。本文除了研究股東對於交易宣告的反應之外，也綜觀不同資產交易方式的現金流量和REITs股利的關連性，藉此瞭解影響REITs選擇交易方式的內涵因素，以及對股票報酬和股利的影響。 / This research examines the tax deferred exchanges made by public U.S. Real Estate Investment Trusts (REITs) over 2003-2006 as well as the transactions of sell-off. The purpose of this study attempts to explore the effects of tax deferred exchange on stock returns and dividend distribution. Result of this study shows that announcement effect of tax deferred exchange is negative in stock value. On the contrary, the relationship between sell-offs and stock value is significantly positive. The reason to explain the difference on announcement effect between two types of property transaction is the specific taxable earning distribution restriction on REITs. U.S REITs have to pay out 90 % of taxable earnings in the form of dividends to their shareholders to exempt from tax. As a result, tax deferred exchange doesn’t bring cash inflow contributing to dividend increase and then shareholders react a lower stock return on tax deferred exchange than on sell-offs. However, the negative effect is weakened by the efficiency of asset reallocation and the regular dividend distribution over tax law restriction. In the analysis of dividend payment, the result of dividend examination supports the hypothesis that tax deferred exchange without cash inflow make dividend fewer than sell-offs. This study may be of importance in explaining the reaction of shareholders on tax deferred exchange of REITs’ property, as well as in providing shareholders with a better understanding of the relationship between cash flow and dividend distribution in order to clarify the cause that affect REITs to utilize different types of transaction and the factors that affect stock return and dividend.

中國大陸自「改革開放」以來，在經濟體制和結構上有著急遽的轉變，同時，海峽兩岸的經貿互動與依存程度也愈來愈高。因此，能有效分析中國大陸經濟情勢的政策模型，對於台灣與中國大陸的政策制定者與政策研究分析者而言皆十分重要。可計算一般均衡（Computable General Equilibrium，CGE）模型由於在數據需求的較大彈性及模型結構上的特性，成為中國大陸現今重要的經濟政策模型之一，本研究的目的在於考察、分析中國大陸可計算一般均衡（CGE）模型的研發背景、發展歷程、建模過程、主要政策應用方向及研究結果。 　　本研究的內容及流程如后：（1）依「歷史研究」、「調查研究」來歸納、分析中國大陸經濟學研究的發展歷程，以及中國大陸可計算一般均衡（CGE）模型的發展歷程；（2）接著依「文獻研究」來分析、分類本研究所蒐集的中國大陸可計算一般均衡（CGE）模型實證文獻（1978年－1998年）；（3）接著依「理論研究」來歸納、分析出可計算一般均衡（CGE）模型的發展、基本結構、數據基礎與建模（Modeling）；（4）以「可計算一般均衡」（CGE）方法，先依據最新的「1995年度中國投入產出表」，編制出兩張：「中國大陸1995年社會會計矩陣」、「中國大陸1995年金融社會會計矩陣」為模型數據基礎，然後應用中國大陸所研發的兩個可計算一般均衡（CGE）模型（「中國大陸經濟-環境可計算一般均衡模型」、「中國大陸金融可計算一般均衡模型」）的主要結構和方程式，以說明中國大陸可計算一般均衡（CGE）模型的建模過程和政策應用方向及模擬結果。 　　本研究的研究結果如后：（1）中國大陸經濟學研究的發展、實際經濟情況變化和政策制定需要，導致中國大陸可計算一般均衡（CGE）模型在「改革開放」後的發展可分為「啟蒙研發」和「政策應用」兩個階段；（2）將中國大陸可計算一般均衡（CGE）模型實證文獻（1978年－1998年）有系統地分類出「貿易政策問題」、「能源和環境政策問題」、「財政和稅收問題」、「經濟改革和發展策略問題」、「外來衝擊問題」、「貨幣金融問題」、「社會保險問題」等七類實證文獻；（3）應用兩個中國大陸研發的可計算一般均衡（CGE）模型，來說明建模（Modeling）與數據編制的過程，並延續、拓展相關的政策模擬研究，分析了環境政策（「綠色導向能源政策」，Green-Oriented Energy Policy：2000年－2015年）與匯率政策（「管理浮動匯率政策」，Managed Floating Exchange Rate Policy：1998年）對中國大陸經濟體的影響。 謝 詞 ii 中文提要 iv 英文提要 vi 中文目次 viii 英文目次 ix 表 次 x 圖 次 xi 1. 緒 論 1 1.1 研究動機 1 1.2 研究目的 3 1.3 研究流程與內容 4 2. 文獻探討 5 2.1 中國大陸經濟學研究的發展歷程 5 2.2 中國大陸可計算一般均衡模型的發展 11 2.3 中國大陸可計算一般均衡模型實證文獻回顧 20 3. 可計算一般均衡模型的發展、基本結構、數據基礎與建模 32 3.1 可計算一般均衡模型的概念與發展 32 3.2 可計算一般均衡模型的一般性基本結構 36 3.3 可計算一般均衡模型的數據基礎與校準 40 3.4 可計算一般均衡模型的計算機求解與建模過程 42 4. 中國大陸可計算一般模型之應用（1）─中國大陸經濟-環境可計算一般均衡模型 45 4.1 模型的基本結構 45 4.2 模型方程式 49 4.3 模型的數據基礎與校準 57 4.4 政策應用-中國大陸宏觀調控下環境政策之一般均衡分析 61 5. 中國大陸可計算一般模型之應用（II）─中國大陸金融可計算一般均衡模型 69 5.1 模型的基本結構 69 5.2 模型方程式 71 5.3 模型的數據基礎與校準 73 5.4 政策應用-中國大陸匯率政策之一般均衡分析 78 6. 結 論 83 6.1 本研究主要貢獻 83 6.2 研究限制與建議 84 6.3 後續研究建議 85 參考文獻 86 附錄1. 94 附錄2. 103 / Both the system and structure of Chinese economy have been changed rapidly since the launch of "economic reform and opening to the outside" in Mainland China, and the economic interaction and trade interdependence between Taiwan and Mainland China are intenser and closer. Effective policy analysis models for Chinese economy is very important to the policy makers and policy analysis researcher both in Taiwan and Mainland China. The Computable General Equilibrium （CGE） model has become one of the most important economic policy analysis model because of its characteristic of higher flexibility on benchmark data and structure. The purpose of this study is to review, investigate and analyze the developmental background, developmental progress, modeling procedure, policy simulations and research results. 　　The brief contents and procedure of this study consist of （1） reviewing on the development of economic research in mainland China and CGE modeling for Chinese economy, （2） reviewing on the literature of CGE models for Chinese economy used on Policy Analysis（1978－1998）, （3） reviewing on the development, basic structure, benchmark data and modeling of CGE model, （4） constructing two Chinese Social Accounting Matrix （SAM） tables developed from the most recent 1995 Chinese Input-Output table to be the benchmark data of two CGE models, the "Chinese Economic-Environmental CGE model" and the "Chinese Financial CGE model", which are developed from Chinese government and researchers, and we revise them in this study,（5） applying those two models to show and illustrate the modeling procedure and results of policy simulations of CGE models for Chinese economy. 　　In conclusion and contribution, it is composed of three parts. （1） The development of CGE model in Mainland China can be divided into two ages, "The Beginning"（1978－1991） and "The Take-off"（1992－）, which results from the development of economic research in Mainland China, the rapidly changed economic system and structure of Chinese economy, and government's urgent demands for policy-analysis tools. （2） The literature of CGE models for Chinese economy used on policy analysis from 1978 to 1998 can be classified into seven groups－ trade policy, energy and environmental policies, public finance and tax reform policies, economic reform and development strategy, external shocks, monetary and financial policies, and social insurance policy. （3） We apply those two CGE models for Chinese economy to analyze the economic and environmental impacts of environmental policy－ " Green-Oriented Energy Policy"（2000－2015） and the economic impacts of the " Managed Floating Exchange Rate Policy"（1998）.