• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 610
  • 548
  • 60
  • 15
  • 1
  • 1
  • 1
  • 1
  • Tagged with
  • 625
  • 625
  • 291
  • 237
  • 191
  • 184
  • 138
  • 120
  • 117
  • 107
  • 106
  • 104
  • 93
  • 90
  • 86
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
251

俄語語言學術語漢譯問題研究-從字典編纂的角度出發

朱逸欣, Chu ,Yi-hsin Unknown Date (has links)
俄語學習者在學習過程中一定會碰到俄語語言學術語。而俄語語言學術語的漢語翻譯存在著許多的問題,容易造成學習者的誤解或誤用。因此本論文透過分析俄語語言學術語翻譯現況和問題來源,並且從術語的特徵來討論術語翻譯時所需注意的要點,探討俄語語言學術語的可譯性,以及實際的術語翻譯手段。 本論文從製作俄漢語言學術語詞典資料庫的角度出發,提供解決語言學術語翻譯問題的方法。透過資料庫的製作,根據資料庫中俄語語言學術語的學科資訊,以及俄語和漢語術語同義詞等資料,分析語言學術語學科分佈現象和語言學同義詞的關係及類型,並且探討語言學同義詞產生的原因
252

卡方檢定在三維關聯結構下之模擬分析與實證研究─以台股原物料族群股價為例

賴宗暘 Unknown Date (has links)
隨著關聯結構方法在1999年開始被應用在財務資料上,對金融市場風險的衡量,可說是一大改革。Dobric & Schmid (2005) Communications in Statistics: Simulation and Computation, 34,pp.1053-1068,提出利用卡方檢定來檢驗二維資料間關聯結構,本文延伸其方法探討卡方檢定應用於三維關聯結構之表現。  首先本文在模擬研究部份,考慮邊際分配未知的情況下,用卡方適合度檢定來檢驗以蒙地卡羅模擬方法模擬Normal關聯結構、t關聯結構、Clayton關聯結構、Frank關聯結構以及Gumbel關聯結構等五種關聯結構。得知隨著切割數的增加,參數估計越來越不精確;而在理論上檢定統計量當樣本夠大時會近似卡方分配,故檢定統計量平均數(變異數)應近似其卡方分配自由度(2*自由度),但隨著切割數增加,表現越不理想;至於檢定力部份,在討論不同情形之下都有不錯的表現。  再之採用台灣股票集中市場中水泥類、食品類、造紙類、橡膠類、運輸類五類股族群,對其日內時間四種頻率:1/9天、1/6天、1/3天、1天的股價報酬率,進行五種關聯結構配適,找出最能夠描述股價日內資料分佈的關聯結構,實證得知上述四種頻率的股價報酬率,皆呈現t關聯結構其自由度為4之配置為最合適。
253

導入資料採礦技術於新巴塞爾協定下企業信用模型-以製造業為例

陳佳樟, Chen, Chia-Chang Unknown Date (has links)
2006 年新巴塞爾資本協定的施行,台灣各家銀行皆陸續步入實施新協定的軌道,發展以風險評等的觀念來計算法定適足資本,讓銀行採自建信用評等系統來評估違約暴險,並透過信用評等達到早期預警的效果,降低信用風險。而資料採礦則是近年來在應用分析領域中相當熱門的議題。 本研究是將資料採礦技術導入企業信用評等模型的建置程序,以國內某一銀行的企業授信資料為實例,資料的觀察期間為2003至2005年,其中針對「製造業」進行研究。藉由企業財務報表簽證資料,結合授信往來紀錄等變數,經誤差抽樣,分別以類神經網路、決策樹及羅吉斯迴歸等採礦技術,建立模型。模型驗證的部份,依據金管會建議的七個方向執行模型之驗證。 研究結果發現,經評估確立以1:2誤差抽樣比例下,使用羅吉斯迴歸技術建模的效果最佳。模型中財務資訊、產業特性及公司的徵審授信紀錄對於違約皆具有一定的解釋預測能力,且顯示產業特性的不同,對於違約機率的預測,有一定的影響。經驗證後,此模型即使應用到不同期間或其他實際資料,仍具有一定的穩定性與預測效力,並且通過新巴塞資本協定與金管會的各項規範,表示本研究之信用評等模型,的確能夠在銀行授信流程實務中加以應用。
254

引用文獻索引資料庫之比較研究 / A Comparison Study of Citation Indexing Database

陳薇竹, Chen, Wei-Chu Unknown Date (has links)
引用文獻索引資料庫在Institute for Scientific Information(ISI)建置了Science Citation Index(SCI)與Social Science Citation Index(SSCI)以後,逐漸為學術界所重視,也帶動了傳統商業公司發展引用文獻索引資料庫之風潮,其中又以Science Citation Index-Expanded(SCIE)及後起之Scopus最為人稱道。但由於傳統商業公司對學者及圖書館收費過於高昂,引起學者及圖書館的反動,興起一陣由計畫及少數商業公司所發展,開放存取引用文獻索引資料庫之風潮,其中又以Google公司製作的Google Scholar,及NEC公司隨著計畫建置的CiteSeer最受人注目。 / 本研究採取實作法為研究方法,評比四個引用文獻索引資料庫的檢索介面及檢索細項之優劣。並以美國計算機械學會(ACM)頒發的杜林獎之50位得主為樣本,對SCIE、Scopus、CiteSeer及Google Scholar四個引用文獻索引資料庫進行作者檢索,逐一過濾檢索結果後,針對正確的檢索結果進行分析,比較四個引用文獻索引資料庫內部重複性與完整性,並交叉比對四個引用文獻索引資料庫兩兩比較之重複性、獨特性及完整性,並歸納造成此研究結果之原因。 / 研究結果發現SCIE與Scopus的檢索方式較容易,不會造成使用者太大的負擔,檢索方式也較為多元詳盡,其中又以Scopus的作者檢索使用最方便;而Google Scholar及CiteSeer皆主要利用一簡潔的檢索列,較難精準的檢索出所需資料。收錄資料完整度方面,Google Scholar收錄資料最多元,SCIE則涵蓋最完整之學術資源。交叉比對結果可得知,Google Scholar之資料獨特性最高;CiteSeer之收錄資料完整度最低。此外除了SCIE以外,其他三個引文索引資料庫皆收錄大量的網路資源。此外,美國計算機械學會的出版品則在四個引文索引資料庫中,皆扮演重要角色。 / 根據研究結果,對此四個引文索引資料庫提出建議,希望傳統商業引文索引資料庫能增加索引網路資源,並調整收費政策;開放存取引文索引資料庫應改正其書目著錄格式;希望圖書館能增加對引文索引資料庫使用之推廣,並教導使用者正確利用開放存取引文索引資料庫。 / 引文索引資料庫索引之文獻,已對學術評鑑造成很大的影響。圖書館應實地使用並引導使用者正確的利用引文索引資料庫,及使用網路資源的正確觀念。如此方可協助使用者不在浩瀚之網路資源中迷失。 / After Institute for Scientific Information(ISI) made Science Citation Index(SCI) and Social Science Citation Index(SSCI), Scholars progressively took notice of citation indexing databases. Commercial Companies also had begun to expand citation indexing database like the famous products are Science Citation Index-Expanded(SCIE). However, the commercial companies charged too much for using the database. So it excited the development of open access(OA) citation indexing database, instant of Google Scholar and CiteSeer. / OA means that people can use these citation indexing database for free. This paper aims to adopt comparison as four databases’ retrieval interface, and unique and overlap of documents of the subjects of computing machinery and electrical engineering. The research subjects are composed of OA and traditional commercial citation indexing database in the follow: SCIE, Scopus, Google Scholar, and CiteSeer. Moreover, this research retrieved all documents of Turing award winners in the four citation indexing databases, in order to examine these four citation indexing databases’ unique and overlap. / As a consequence, this study provides the findings as follows : Firstly, traditional commercial citation indexing databases (SCIE and Scopus) have the easier retrieval interface and various searching forms. The Google Scholar collects more multiform resources of retrieval results, and SCIE completed collects scholarly literatures. We make a comparison to find that Google Scholar has much more unique data, but CiteSeer is completely less in four citation indexing databases. Besides SCIE, another three citation indexing databases conclude a large number of internet data. Finally, publications of The Association of Computing Machinery(ACM) play an important role in the four citation indexing databases.
255

金控合併銀行標的選擇與經營效率之關聯

田蕙綸, Tien, Hui-Lun Unknown Date (has links)
本研究利用資料包絡分析法探討被併購標的銀行於購併前後是否能增加效率,以及具有何種特徵之銀行較易成為金控公司併購之對象。 研究結果發現標的銀行於購併之後,雖能促進效率,但結果並不顯著,顯示合併之後效率提升是綜合標的銀行之高資產品質、高獲利能力、低利率風險及分行數少才使購併後效率提升。在標的銀行的特徵方面顯示當銀行處於非遞減規模報酬、財務上呈現弱點,但銀行資本結構較佳、利率風險較低時,容易成為金控公司購併的對象。 / This paper tried to apply Data Develop Analyis (DEA) to examine whether M&A could promote the targeted banks’ efficiency and the criterions when financial holding companies choose the targeted bank. The experical resoults suggested that although the targeted banks denoted higher efficiency gains, however, the gains do not achieve statiscally signicant level. The results implied that the higher efficiency gains might come from the targeted banks’ high quality assets, greater profitability, lower interest risk and less number of branches. On the other hand, the banks operated under non-decreasing return to scale, financially vulnerable banks, better capital structure, and lower interest risks were more likely to be the acquisition targets.
256

Basel II下應用商業智慧技術於個人信貸信用評等模型之建置

曾詩軒 Unknown Date (has links)
新巴塞爾資本協定(Basel II)之內部評等法(IRB)的運用關鍵,在建立一個有效的信用風險模型,而此模型的功用在於將銀行的風險程度,以量化的風險因子來表達;而本研究即是針對新巴塞爾資本協定,探討在新協定的最低要求(Minimum Requirement)下,銀行欲採用內部評等法(IRB)架構信用風險系統時,應如何建立信用風險模型中「違約機率(PD)」的量化流程。 本篇研究以國內某家金融機構的資料為例,利用羅吉斯迴歸來進行製作信用評分模型,在信用評分模型正確性的指標測試中,不論是在Kolmogorov-Smirnov值、ROC比率、Gini係數的測試上,皆比原此家金融機構在正確性指標測試中更為出色。 最後,更進一步依照該模型所預測之違約機率,建立信用評等,並同時探討不同等級之客戶特性,使金融機構能更有效率地加強其風險控管,進而改善其顧客關係管理系統。
257

改良式強迫選擇量表應用於甄選情境及其對應徵者作假傾向的影響

史尚軒 Unknown Date (has links)
強迫選擇型式量表最大好處為防止填答者作假,但若運用於甄選情境,勢必要面對量表結果是否能作個人間比較的問題。 過去學者認為強迫選擇型式量表不能作個人間比較,主要是因為其測得的向度彼此間互依(亦即向度間呈現負相關),因此其產出的資料只能反映個人特質的相對強度訊息。本研究認為若能讓向度間的互依性降低,就有可能測得個人特質的絕對強度,本研究依此為目標,試圖運用改變計分方式來降低量表互依性,主要以五大人格因素中的外向性與勤勉審慎性為架構,編製出改良式的強迫選擇量表,期望其在甄選情境下:1.產出的資料可以測得個人特質的絕對強度;2.比李克特氏型式量表測得更準確的個人特質;3.比李克特氏型式量表更能防止作假。 本研究採模擬實驗法,參與者被要求在網路上模擬應徵汽車業務員的工作,研究者操弄兩個獨變項:作答情境(甄選、自我瞭解)以及量表型式(改良式強迫選擇型式、李克特氏型式),皆採受試者間設計,依變項為外向性、勤勉審慎性之原始分數及z分數、社會期許分數,並以104人力銀行的職能性格量表當作情境間比較標準。總共蒐集559份有效樣本,研究結果發現,在甄選情境下,改良式強迫選擇量表能和自我瞭解情境下104職能性格量表測得相當的外向性分數,也較李克特氏型式量表更能防止參與者作假,但是並無明確的證據顯示在甄選情境下,改良式強迫選擇量表比李克特氏型式量表更能準確地測得個人特質。 在勤勉審慎性部份,改良式強迫選擇量表在甄選情境下與104職能性格量表測得的結果相關甚低,同樣較李克特氏型式量表更能防止作假,但亦未能顯示改良式強迫選擇量表在甄選情境下,比李克特氏型式量表更能準確地測得個人特質,研究者針對此結果進行較深入的討論。
258

加入WTO後中國大陸產險市場經營效率之研究

許仁榮, Hsu, Jen Jung Unknown Date (has links)
中國大陸於1986年申請重新加入WTO,經過15年的長期努力及多方談判,並且在各種產業上做出重大讓步之後,於12月11日正式成為第143個正式的會員國。加入WTO之後,在大陸金融市場規模龐大且成長迅速的吸引之下,市場的開放必然會吸引大量的國際保險業者進入大陸。另外,與一般開發中國家的情況相似,大陸的金融產業相對於其他產業而言是發展較為遲緩的產業,金融機構進駐後造成的衝擊必然十分嚴重,因此大陸當局應儘早加以準備與因應。 本研究採用資料包絡分析法進行效率值之計算,並使用大邊界法建構出單一效率邊界,探討大陸產險業在加入WTO前後的效率變動情形。本研究的結果如下:1.根據單變量無母數統計分析,整體產險業在加入WTO前後三個效率值雖呈正向關係但不顯著,表示階段性的開放是較理想的措施。2.非中資公司的整體技術效率值、純技術效率值與規模效率值均高於中資公司,顯示非中資公司其全球化的經驗較中資公司更好、更成熟。 / Mainland China reapplied in 1986 to join WTO, passed through 15 years long-term endeavor and negotiates in every way, and made the significant concessions after each industry, on December 11, 2001 officially became the 143rd official member nation. After joining the WTO, the mainland's financial market scale and rapid growth attract, opening of the market will attract a large number of international insurance industry to the mainland. Furthermore, with most developing countries in similar circumstances, the mainland's financial industry compared to other industries in terms of development is relatively slow industries, financial institutions stationed after the impact is inevitable, therefore, the mainland authorities should be prepared as early as possible and to respond. This research applied Data Envelopment Analysis (DEA) to measure efficiency scores, and applied Grand Frontier to measure the trend of the efficiency change before and after joining WTO. The results of this study are as follows: 1. Based on the univariate statistical analysis, overall in the property insurance industry after joining the WTO three efficiency values have shown a positive relationship but not significant, said the gradual opening is a relatively ideal measures. 2. Non-Chinese company's overall technical efficiency, pure technical efficiency and scale efficiency were higher than the value of Chinese-funded companies that show non-Chinese-funded companies in their global experience are better and more mature than those Chinese-funded companies.
259

NBA外籍球員薪資與效率衡量之關聯性研究

王浚宇 Unknown Date (has links)
本研究採用資料包絡分析法球員效率與NBA官方網站效率值來探討其對球員薪資之關聯性。本研究除探討外籍球員與美國籍球員間效率及高薪球員與低薪球員間效率差異,並同時比較資料包絡分析法(DEA)所計算之效率值與NBA官方網站所提供之效率值何者與薪資之關聯性較高。 實證結果發現,外籍球員與美國籍球員效率無顯著差異;但是高薪球員的效率顯著的優於低薪球員。另外也發現薪資與國籍間之關聯,及國籍與效率間之關聯性並不顯著。最後,在Vuong test比較資料包絡分析法之效率與NBA效率值與球員薪資間之關聯性的測試,雖然兩類效率值前一期與薪資之關聯性都成正相關,但是DEA所計算之前一期效率值高於NBA官網之前一期效率值與薪資關聯性,顯示DEA所計算之效率值可以比官網之效率值更能做為球員薪資談判之指標。 / This research applies Data Envelopment Analysis and examines the relationship between NBA players’ salaries paid and efficiency performance. The research not only compares the efficiency difference between players paid for different salaries, Local players and Foreign players, but also construct a model to examine the explanatory ability about player’s salaries between DEA efficiency scores and NBA official efficiency scores. The empirical results suggest that there is no difference between the local players and the foreign players, but the efficiency performance of high-paid players is better than the low-paid players. After we controlled the other variables, the empirical results illustrated that player’s nationality and salaries has no significant influence in player’s efficiency. The Vuong test results reveal that the pervious period DEA efficiency score demonstrated a higher explanatory ability than the pervious period NBA official efficiency score. The finding suggests that the DEA efficient approach could be a better index for players’ salaries negotiation than NBA official efficiency score.
260

評估我國大學校院之發展策略:以資料包絡法為例

傅遠智, Fu,Yuan Chih Unknown Date (has links)
本研究旨在運用資料包絡法(data envelopment analysis, DEA)分析「教育部公私立大學校務資訊評鑑系統」中90至92學年度我國50所大學校院在研究與教學上之經營效率。研究問題有四:(1)國內大學校院教學與研究之投入與產出性指標系統分別為何?(2)國內各大學校院之相對效率值為何?改進策略為何?(3)國內各大學校院之發展類型為何?(4)國內各大學校院之發展趨勢為何?本研究考量投入項與產出項之重要性,編製「評估我國大學校院之發展策略指標權重專家調查問卷」以國內30位高等教育行政主管為調查對象進行層級分析取得指標相對權重,並套用於確定區域模式進行效率值之估算。指標係依據文獻探討與積差相關的同向性檢定,在研究效率上選擇6個投入項及5個產出項;在教學效率上選擇7個投入項及2個產出項。所使用之統計分析包括:積差相關、層級分析法、資料包絡法(效率分析、差額變數分析、參考集合分析、Malmquist 生產力指數)、Kruska-Wallis H檢定、Mann-Whitney U檢定、Tobit迴歸。 本研究得到以下六項結論: 一、本研究依大學生產關係架構所擬定之20項指標經相關分析、專家權重以及資料包絡法三種層面之分析,可作為以效率觀點評估我國經營效率的指標系統。 二、研究投入資源重專任副教授以上人數,研究產出重專任教師發表重要期刊論文篇數;教學投入資源重專兼任教師人數,然而,授課時數與約當畢業生人數在教學產出上同等重要。 三、三個學年度間,研究面向中有5校均為整體技術有效率,40校均為整體技術無效率;教學面向中有1校均為整體技術有效率,48校均為整體技術無效率。研究與教學無效率之原因均導因於純技術無效率,分別約浪費22.4%-27.2%以及26%-27.3%的投入資源,規模無效率影響較小。 四、不同權屬別對於整體技術效率之影響,主要源自於純技術效率;不同學校規模對於整體技術效率無影響,其係綜合學校規模越大純技術效率越差,但相對規模效率越佳的結果。 五、國內各大學校院發展,研究型大學以國立大學與醫學校院為主;教學型大學以師範校院與私立大學為主。 六、就三個學年度發展趨勢,整體而言研究效率提升,教學效率衰退,而各大學發展趨勢除整體技術效率略有消長外,35校在三個學年度發展類型均為一致。 / The purpose of this study is to assess the research and teaching productivities of 50 Taiwanese universities and colleges over the period of academic years 2001-2003. Research questions include: (1)What is the index system to assess school performance? (2)How many efficiency scores do these schools get? Inefficient units can implement what kinds of improvement to achieve efficiency? (3)What are their development styles? (4)What are their development tendencies? In order to consider weight restriction on inputs and outputs, author compiles “the expert questionnaire of assessing development strategy in Taiwanese universities and colleges”, and invites 30 specialists to answer, including 15 principals in higher education institutes and superintendents from Ministry of Education and 15 professors majoring in this field. Assurance region model with weigh indexes is used to calculate the whole efficiency scores. According to the result of literature reviewing and isotonicity test, author picks out 6 inputs and 5 outputs in research dimension and 7 inputs and 2 outputs in teaching dimension. Some statistic methods are used, including Pearson correlations, analytic hierarchy process, data envelopment analysis, Kruska-Wallis H test, Mann-Whitney U test and Tobit regression. Results based on data show as follows: 1. Through statistic analysis, those indexes the author drafts can be used to assess the productivity of Taiwanese higher education institutes practically. 2. In research dimension, the most important input is the number of assist professor, output is academic paper accepted by prestigious journals. Besides, in teaching dimension, the most important input is the number of faculty; however, the amount of class hours is equal to the amount of graduates on output part. 3. In research dimension over three academic years, 5 schools were efficient in each year; however, 40 schools were inefficient in each year. Furthermore, in teaching dimension, only 1 school was efficient in each year; in contrast, 48 schools were inefficient. Both of research and teaching inefficiency mainly came from pure technical inefficiency, wasted 22.4%-27.2% and 26%-27.3% separately. 4. The effect of different authorities upon the efficiency mainly came from the pure technical efficiency. In addition, because of the effect mixed pure technical efficiency (PTE) with scale efficiency (SE) the effect of the different school scale upon the efficiency was non-significant. 5. Research universities are mostly from national universities and medical colleges. In contrast, teaching universities are mostly from private universities and normal colleges. 6. As a whole, research efficiency score grows but declines in teaching. Beside of some fluctuations on efficiency score, there are 35 schools maintain their development styles stably.

Page generated in 0.0389 seconds