分析具有巢狀誤差結構的迴歸模式時，惹忽略隨機誤差項之間的相關性，而採用最小平方(OLS)估計量所導出的標準 F 統計量（以 F^S表之）進行檢定，會導致過大的型 I 錯誤機率；若將隨機誤差項之間的相關性納入考量，而採用廣義最小平方(GLS)估計量所導出的 F 統計量 （以 F^GLS表之），則計算上會較為繁雜。因此我們藉由轉換方式，將模式轉換成隨機誤差項之間彼此獨立的新模式後，再以 F^S 進行檢定，其結果與直接以 F^GLS 檢定相同，且可使計算較為方便。由於模式轉換所需的轉換矩陣為母體變異數的函數，因此當母體變異數未知時，我們以 Henderson 的常數配適 （fitting-of-constants）方法來估計之。藉由模擬結果得知，若各段的觀察個數相等，則不論巢狀誤差結構為二段式(two-stage)或三段式(three-stage)，廣義最小平方估計量(GLS)均較最小平方估計量(OLS)表現穩定，且 F^GLS 在檢定力及實際顯著水準方面的表現也都比 F^S 好。 / When analyzing the regression model with nested-error structure, if the correlations between errors are ignored, and conduting the model adequacy test by the standard F statistic (F^S) led from the ordinary leastsquares estimator (OLSE) , then the type I error rate will be inflated. However, if the corrlated structure is considered and the model is tested by F^GLS led from the general least-squares estimator (GLSE) , the calculation will be more complicate. The model can be transformed to a new model with independent random errors and then, tested by F^S . The result is the same as the one by F^GLS , also it is more convenient for calculation. Since the transformation matrix is a function of variance components, we estimate variance components by Henderson's fitting-of-constants when they are unknown. Through simulation, it is concluded that if the observations in each stage of nested-error structure are the same, the GLSE is more stable than the OLSE in both two-stage and tree-stage structures. Also, the power and the sizes of F^GLS will perform better than those of F^S .

次級房貸風暴使得包裝複雜的衍生性金融商品紛紛遭受波及後，目前結構型金融商品的條款設計將朝簡單化和透明化的趨勢發展，有助於全球金融市場的效率性、完整性與穩定性。本文從市場上選擇具代表性的利率結構型商品，應用模型來推導商品的價格，並深入分析商品的報酬與風險型態。 本文分析的個案商品為全球知名的匯豐銀行所發行之十年期「固定期限交換利率利差連動債券」，在評價上將採用LIBOR市場模型，利用市場上既有的資料求算出期初遠期利率，並校準模型所需的參數化波動度函數與相關係數函數，建立與市場一致的利率期間結構與利率波動度期間結構。模擬路徑時應用最小平方法蒙地卡羅來求得該商品發行之期初價格，此外，亦採用反向變異法加速收斂效果，並針對商品的條款設計作拆解與分析。最後，本文探討了發行機構發行商品之風險與避險策略，並且從投資人之報酬及風險層面作詳盡地剖析。

90年代的後期，敏捷式系統發展方法開始被倡導。相對於傳統的系統發展方法，敏捷式系統發展方法著重於回饋機制而非事前的計畫、以人為中心而非以流程為中心。這樣的方法希望能助於提高組織對回應市場、客戶的效率，進而提高效益。目前在商場中使用此方法做為開發工具的企業仍是少數，本研究希望透過探討敏捷式系統發展方法論的使用時機來進行教學，進而得知系統開發人員對於接受敏捷式系統發展方法的關鍵因素，並藉此了解該如何在企業中導入此方法。本研究發現除了使用此方法的能力會影響影響使用意圖之外，在內在因素方面也包含了公司結構與團隊因素，外部因素則包含了顧客與成功案例因素，這些都是接受敏捷式系統發展方法的關鍵因素。本研究希望根據以上的分析結果，提出敏捷式系統發展方法導入之建議，提供組織做為參考用。 / Awareness of agile system development methodologies (SDM) has grown among information systems development community in recent years. Many of their advocates consider the agile and the plan-driven SDMs polar opposites. Indeed there are circumstances where agile SDMs are more suitable than plan-driven SDMs. Yet, there have been few studies on understanding developers’ adoption intention. This paper takes an initial attempt to gauge new generation of software developers’ intention to select agile SDMs. To many of these developers, agile SDMs are relatively new if not unheard of, in order to assess their intention to choose such category of methodologies, this research first introduced the methodologies to a group of 21 IS-major graduate students and discussed how and when to use agile SDMs. Then a survey was conducted, which was comprised of two parts of questions: agile SDM self-efficacy and intention to use. PLS analysis results showed that agile SDM self-efficacy influence the intention to use through performance outcome expectation, personal outcome expectation, and affect. Although the relationship between self-efficacy and anxiety was not confirmed, anxiety does affect intention to use. The fact that direct relationships between all four emotive variables and the intention to use are established implies that in order to encourage the use of agile SDMs, the focus should be emotive variables, and that self efficacy may be just one of various ways to promote the favorable emotional states. In addition, these participates were invited to a three-round Delphi test and analytic hierarchy process to retrieved their concerns about accepting or rejecting agile SDMs. Ten key factors were extracted and categorized. Adding up the pros and cons, team dimension is the most important dimension, which explains individual first concerns about how the collaboration when using agile SDMs. Other than team dimension, customer, corporate structure, project, success cases and methodology dimensions were consistent with the literatures. Thus our study provides a critical understanding of the factors that affect new generation of software developers’ intention to select agile SDM.

微陣列晶片是發展及應用較為成熟的生物晶片技術。由於微陣列實驗程序複雜，故資料常包含多種不同來源的實驗誤差，為了適當的區分實驗中來自處理、晶片及基因的效應，我們提出混合效應變異數分析模型來調整系統誤差。針對各基因在不同實驗環境的差異性假設檢定問題，利用最小平方法推導出點估計以及對應的檢定統計量。本研究介紹多重檢定問題中的族型一誤差，並證明在此模型下，Sidak調整法為適當的多重檢定方法。在給定族型一誤差率的顯著水準，利用檢定力的公式，運算出在預設檢定力的最低水準下所需最小樣本（晶片）數。最後我們透過電腦模擬，以蒙地卡羅法來估計檢定力與族型一誤差率，由模擬結果發現，採用此最小樣本數結果，其檢定力可達到預期的水準以上，並且其族型一誤差率皆適當地控制在顯著水準以內。

近年來由於溫室效應加劇以及氣候變遷加劇，因此符合綠色運輸特性的公共自行車系統，成為各國交通部門發展綠運輸政策時的目標之一，同時，大數據分析亦是目前受到高度關注的熱門議題。而本研究首先使用臺北市微笑單車租借大數據探討在不同時間點下民眾日常使用微笑單車之旅運行為，分析不同站點間的旅次特性。再運用社群網絡分析，以站點之間旅次連結多寡作為權重，探討站點間之緊密程度，以及不同時間點下微笑單車租借量之熱點分布情形，並將其視覺化呈現。 後續透過文獻分析，擷取影響公共自行車使用量之因素後，本研究嘗試運用一般線性迴歸模型與地理加權迴歸進行模型建立，並探討各影響因素對於旅運需求之影響情形。實證結果顯示，地理加權迴歸模型可以解決一般線性迴歸所產生空間自相關問題，使得模型解釋能力獲得改善。本研究並使用地理加權迴歸進行使用需求分析以及預測，對未來公共自行車營運以及站點擴張提出結論以及建議，期能提升公共自行車系統之使用量。 / Due to the climate change and aggravation of the greenhouse effect in recent years, the public bicycle system with the feature of low-carbon emission has raised more and more attention internationally, and has become one of the targets in developing green transportation policies of transportation departments of governments around the world. Meanwhile Big Data analysis issues, on the other hand, are currently a sought-after topic which has caused great concern as well. In this study, we utilize the rental data of the YouBike system in Taipei to discuss the public usage of YouBike tour at different periods. With the use of social network analysis, we discuss the relationships between different bicycle stops based on applying the number of travels between different sites as the weight. Eventually, the hotspot analysis will be carried out by operating the GIS system. In this way, we are able to discuss the hotspot distribution of YouBike rentals in different time and then visualize the result. After that this study pick up the variables which will effect the YouBike usage by reference review. This research try to built models by utilizing the Least Squares Method and Geographically Weighted Regression. Then we will have a discussion with the result of the two models. The result shows that Geographically Weighted Regression can resolve the spatial autocorrelation problem which happened in the Least Squares Method and to gain a better result. With the analysis and prediction of public bicycle system from Geographically Weighted Regression, we hope to raise the usage of public bicycle system by concluding as well as making recommendations for the future operation of public bicycle and the expansion of bicycle stops.