我國為順應國際自由化之潮流，並發展台灣成為亞太營運中心及加入世界貿易組織，師法先進國家之經驗，逐步推動電信自由化，網際網路業務已於八十四年開放，可說是網際網路商業化應用時機的已經來臨；因此針對網際網路業務特性、現行價格訂定之瞭解及檢討，進而針對企業之需求，考量其服務之改進，可預期對今後我國網際網路業務將甚具參考價值。 　　本研究主要分二大部份，第一部份是網際網路使用基本分析，主要是根據問卷調查法，及業者的實際使用情況，來分析網際網路的使用型態，問卷寄發對象分成固接用戶及撥接用戶二種，固接用戶之寄發對象係以HiNet全球資訊網於84年12月，工商服務欄中之企業用戶124戶全數寄發；撥接用戶是由HiNet之經銷商，隨機取得撥接用戶寄發200戶，固接用戶回收率為18%，撥接用戶回收率為14%。研究的第二部份建立－模擬分析模式，來預估在各種不同的需求預測、服務指標、費率結構時，網際網路服務之營收、成本的變化及其損益平衡之分析，以作為網際網路資費訂定策略之參考。 　　本研究所得之結論可歸納為下列四點： 　　一、網際網路使用者分析：有八成的撥接用戶其引進過程是由報章雜誌得知，近八成的固接用戶其引進過程則是由同業介紹；撥接用戶使用時間在一年內者佔79%，固接用戶使用時間在一年內者佔45%，且86%的受訪企業用戶其研發部門均使用網際網路；撥接用戶與固接用戶使用率最高的服務資源均為WWW，並希望網際網路具傳真之功能；撥接用戶將應時間慢」視為最重要的問題，因接用戶將「安全性的顧慮」列為最重要的問題。 　　二、使用費率分析及滿意度分析：撥接用戶大致上認為HiNet費率屬便宜，固接用戶則認為基本費、國際通信費，這些費用均偏貴；撥接用戶及固接用戶同對網際網路最滿意的項目為對外連繫方便；撥接用戶選擇網際網路服務業者(ISP)的第一優先考量因素為費率較低及連線服務廣泛，而固接用戶第一優先考量的則為連線品質。 　　三、使用型態分析：撥接用戶每日的清晨3至10時，為明顯的離峰時段，夜間九時起開始步入尖峰時段至十二時左右；而固接用戶使用者在工作日時使用量較週末明顯為高；每日的下午1至5時為尖峰時段。 　　四、模擬分析結果：以85年5月為基準點(1)在高成長趨勢下現行費率結構、維持目前之服務等級(C)時，往後預測約第二個月後，可達損益平衡點。(2)高成長趨勢下差別費率、維持目前之服務等級(C)之情形，在預測的第二個月，開始有1%的利潤。(3)高成長趨勢下，差別費率結構、較佳之服務等級(B)之情形，於預測開始的第六個月後可達損益平衡。(4)高成長趨勢下差別費率結構、更佳之服務等級(A)時，於預測開始後的第12個月仍有19%的虧損。(5)中度成長趨勢下，差別費率結構、服務等級(C)時，在預測開始後的第三個月，可達損益平衡。(6)中度成長趨勢下，差別費率結構、服務等級(B)時，於預測開始的第九個月達成損益平衡。 　　在撥接使用者數達到約10萬戶，固接用戶達到約600戶時，維持目前HiNet的服務水準與費率策略，即可達損益平衡點；為增加收益，可採取離峰時間差別費率策略。但為提昇經營形象及連線品質，以長期觀點來看，似應將目前的服務等級C提昇至服務等級B，於撥接使用者數達到約18萬戶，固接用戶達到約750戶時，並採取離峰時間降價促銷策略，即可達損益平衡點。

每月公告營收額實施由來已久，過去文獻之研究，其實證結果大多具有資訊內涵，惟該資訊之允當性如何及其與會計師簽證數之差異如何均未曾加以探討。本研究之目的乃在探討公告營收額與會計師簽證數間之差異狀況，並試圖尋求解釋該差異之因素與解釋模型。 　　本研究主要分為二部份：一為目前差異情形之分析，一為假說之檢定。文中以代理理論為基礎，導出董監事持股比例、經理人持股比例、長期負債權益比、總資產金額、已上市年度數、以及簽證方式等六個代理變數，並加入產業變數進行分析，以尋求最佳之解釋模型。 　　研究結果顯示：（一）敘述性統計顯示實施查核程序之調整金額常高於執行核閱程序之調整金額，在調整金額之方向則調低營收額之次數高於調高營收額之次數，而調整門檻粗略估計約為10萬元左右，調整金額之百分比大多在簽證數之20%以下。（二）迴歸分析實證結果得知：(1)董監事持股比例與差異數有顯著關係，但董監事持股比例愈高，差異數愈小。(2)經理人持股比例與差異數幾乎無關，其原因可能為經理人持股普遍太低，操縱營收額可能對其持有或出售公司股票獲利之誘因不大，也可能是因為營收額之允當與否，代表經理人是否盡了代理人之職責，若有太大差異可能會影響其報酬。(3)長期負債權益比與差異數幾乎無關，其原因可能為債務合約關注之重點為淨利、流動比率、利息保障倍數、抵押品等因素，而對營收額之允當性可能沒有規定或監督。(4)公司總資產額愈大，差異數愈大，此與假說相反，其原因可能為公司愈大，其業務較為複雜，以致產生差異之機率增高，再者公司設置監控系統之成本在目前國內證管會之要求下已固定，且其成本於上市公司並非重大，以致考慮成本效益之情形大為降低。(5)公司上市愈久，差異數愈小，此與假說相符，換言之，公司之內部控制愈好，差異愈小。(6)簽證方式不同，差異數不同，查核之差異高於核閱之差異，此與假說相符。

網路廣告是目前世界上成長最快的廣告類別，每年營收以30％的速度在成長，是最具有發展潛力的廣告市場。但是對於廣告主而言，如何評估網路廣告的效益及其收費模式發展，與未來廣告市場的發展息息相關。因此，本研究希望藉由網路廣告的型態及計價方式探討，試圖找出發展的趨勢，提供給業者以及學術研究作為參考。透過本研究，讀者可以大致瞭解目前網路廣告的發展狀況，進而對網路廣告的發展趨勢也有所瞭解.。 目前網際網路的型態大致有：關鍵字搜尋、展現式、分類目錄、導引轉介、電子郵件、置入性行銷等六大類型。而網際網路的廣告收費模式，一直在不斷發展中，從傳統的記錄曝光次數來收費，逐漸轉入網路上專有的依點閱計費費的方式。本研究透過此一趨勢，經過了資料整理分析研究後，本研究得出結論，網路廣告的營收模式方向，朝著「使用者提供之資訊含量度增加」以及「廣告計價方式之自由度增加」兩個面向來發展。 所謂的「使用者提供之資訊含量度增加」，代表著因為網路互動的特性，透過資訊科技來收集使用者在觀看內容時所提供的資訊，藉由分析這些使用者提供的資訊，就可以針對使用者資訊及廣告內容之間的關連性，對廣告做更有效率的投放。如此一來，廣告效果大幅度增加，相對而言廣告的價值也隨之提升。 而「廣告計價方式之自由度增加」，由原本的賣方市場，逐漸轉為由買方競價之市場。也就是說透過自由市場的機制，真實呈現廣告的價值。使用者更可以藉由資訊科技的幫助，隨時觀察網路廣告的狀態。在價格透明、資訊公開的結果下，間接的促進了網路廣告市場的成長，也確立了未來廣告定價的發展方向。 本研究同時也提供了「網路廣告代理制度的建立」，以及「無線廣告之經營模式」，等兩項的發展建議，提供給網路廣告業者及電信業者參考，此部分亦深具研究價值，期待未來能有更多的研究報告出現。 / Online advertisement is the fastest growing kind of advertisements, and already attracting notice from all over the world. Its revenues increased about 30% per year，thus is the most potential advertising market. Evaluating the effective and price model of online ad are to be closely bound up with it future development. This thesis focuses on the type and price model of the online ad, and tries to find out the trend of development. This is for business and learning reference. Through this research, the reader will get to know the situation of online ad development, further more to understand the trend. The types of online ad are classified into six types: keywords, display ad, classified, lead generation/referral, and email ad. However, the pricing model is still in progress. From the traditional exposure and impression model transform to by click and action model. Our research discovered this phenomenon. Through comprehensive data collection and analysis we conclude that the online ad is heading to two directions, “the increase of user information provided” and “the flexibility of ad price”. “The increase of user information provided” means we can collect more and more information from user’s action through browser. With the information technology we can have a lot of analytic works to know what user want, and properly deliver the ad that highly relates to what users want. .It will make the online ad more effective result of increasing the value of ad. “The flexibility of ad price” means the internet market turning from sellers to buyers. The efficiency reporting system provide advertiser instant statistics, which makes the value of online ad effectiveness transparent to customers. Indirectly lift the growth of the online ad market and also affirm the pricing direction of online ad. Our research also provided some suggestions for ad agents and mobile ad development for online ad or telecom business operators, and we hope it will be more valuable for these types of business.

過去電信事業強調無論任何一端用戶透過電話機、傳輸線路、交換機等網路連接至另一端用戶，進行點對點(end to end)語音或數據之通信時，基於網路不可分割之特性，進行整體固定網路設備投資相當龐大，且因具規模經濟特性而形成自然獨占；加以其多屬民生攸關之公用事業，又須具備一定之服務品質及可負擔之合理價格，始能提供不論居住何處之全國民眾使用，故大多由政府主導經營。 　　然隨著電信自由化來臨，民國八十五年七月一日電信解制，分為監理電信業務之電信總局及國營事業之中華電信公司，確立監管架構及電信業務開放以引進競爭市場。惟為避免新進業者會對更多需求及更可獲利的城市地區首先提供服務，而忽略了鄉村偏遠地區，形成吸脂作用(Cream skimming)。故顯然電信自由化後，儘管商業利益的考量是競爭市場存續的重要因素之一，然而為避免關切公共服務的精神隨之而消失，那麼政府如何適度藉入以求得其間的平衡性，這應屬必要的。 　　基本上，應以國民基本通信之概念考量普及服務之範圍，同時應基於公平及合理的原則下，以負擔得起且日常生活中有必要的電信服務；然由於在公平的、合理的、負擔得起、有必要性等充分條件下，仍須加上一定的「服務品質」為必要條件，才能使普及服務的精神顯露，特別在隨著電信科技發展、社會環境變遷及民眾需求，仍須作階段性不同之定義，才能符合普及服務的真正目的。 　　因此，本論文採用文獻探討、政策德菲法(Delphi method)及實例探討，從管制理論的基礎，說明管制因環境之所需，尤其當電信自由化之初，引入市場競爭之同時，適時的管制對於公平的競爭是有需要的；另再參酌美國、英國、澳洲、香港及法國等先進國家之電信普及服務機制，透過美國、英國、澳洲及香港等國家之電信監理機構之網站，及相關電信自由化之文獻，蒐集到各國對普及服務相關之管理方式與法規規定，歸納出各國對普及服務目標之界定，普及服務之提供者，成本計算方式，分攤普及服務之義務者及分攤方式等，再進行整理、分析與比較，嚐試提出一個適合國情的普及服務制度，並對實務執行所面對的問題提出建議，以供主管機關參考。 　　本論文結論重點為，除建立一套含理論及實務運作的電信普及服務機制外，為發現整理出相關問題如：(1)「電信事業普及服務基金」非為政府基金之管理，是否須要編列年度業務計畫及預算書，值得商榷。(2)就商業利益考量，固網業者收取申裝市內電話之「界外工料費」應無不可，然是否能適用普及服務補助的問題。(3)易將普及服務基金視為另一個社會安全或防災救助的課題，扭曲其設立之精神。(4)提供普及服務所產生之無形利益是否應納入棄置營收之計算。(5)海岸電台船舶遇險及安全通信服務是否納入普及服務之考量。 　　其政策建議重點則為：(1)為消彌外界之疑慮，並兼顧效率運作之原則，除必要之管理費用應撙節使用外，相關之收支帳目宜經由會計師簽證予以公開透明化。(2)為維護消費者之權利，裝置材料費用應有下降趨勢，應重新檢討並提出合理成本，訂定合理之收費方式。(3)社會安全或急難救助所經費，應由社會福利安全政策編列預算支應，以免增加電信業者之成本負擔，並轉價至最終消費者。(4)無形利益應可加以計算納入棄置營收項目內，可利用每一家或前五大業者平均的廣告效果估算；或利用消費者行為之抽樣問卷調查，求出潛在消費者及平均消費之貢獻，做為無形利益的推估。(5)可成立專案小組，針對海岸電台設備及維運費用等訂定效率化之衡量標準；或以公開招標方式，徵求自願投入之電信業者；或協調海巡暑整編至該單位。(6)必須建立一套成本資料管理及稽核制度，以增加核算資料的明確性。(7)考慮修法開放可以行動電話做為普及服務之提供，公開徵詢有興趣之行動電話業者，並於北、中、南區各找一處以行動電話提供普及服務之實驗。(8)第一類電信事業部分，可由普及服務管理委員會就經營效率性、虧損之持續性等相關因素，予以審核後得免予繳交；第二類電信事業部分，以經營語音單純轉售及網路電話兩項從事語音電話服務為主，衡酌經營差異程度，排除一定營業額以下者可免予繳交。 / In the past, in order to provide seamless end-to-end communications services for the end users, telecom enterprises that required cost-sunk expenditure, economies of scale and operated an indivisible network had been regarded as a naturally monopolistic business. Moreover, telecommunications as one of public services relevant to the people’s livelihood has to be provided for all nationals at an affordable price and a reasonable level of service quality. As a result, most telecom enterprises worldwide were operated and conducted by the Government. 　　Nevertheless, along with the global trend of telecom liberalization, on July 1 1996 our government reorganized the DGT into a sole telecom regulator and established the state-own holding company, CHT, to provide telecom services, with a view to build up a regulatory regime and to open the telecom market further. To avoid newcomers in a competitive environment tending to focus on high revenue growth telecommunication markets, such as metropolitan areas, resulting in the phenomenon of ‘cream skimming’ and in the impediment to the spirit of public services, there is a need for government intervention in the provision of telecom services. 　　Basically, the concept of universal telecommunications service shall be confined to the basic telecommunication services which are indispensable to every citizens and should be provided on reasonable request at an affordable price and a reasonable level of quality. However, universal telecommunications service might has different definitions over time so as to fulfill its policy objectives in each phase given the development of technologies and the changes in social environment. 　　Accordingly, by conducting Documentary-type method, Delphi method and Case Studies, this dissertation has illustrated with regulatory theory why regulation shall be based on the requirement of the environment. Particularly, during the early stages of telecom liberalization, regulatory involvement at right moment is required for creating a level-playing field when introducing competition. From the websites of several telecom regulators in advanced countries, e.g. US, UK, Australia, HK, etc., and some literatures on telecom liberalization, this dissertation has collected those countries’ management criteria and legal framework for universal service and summed up their universal service objectives, providers, cost calculation methods and obligation sharing mechanisms. Then through a further analysis and comparison, it has tried to propose a universal service mechanism that might better fit our national conditions and offer the authority concerned some practical recommendations for the implementation issues. 　　In conclusion, this dissertation has built up a universal telecommunications service mechanism based on both theoretical analyses and practical considerations. Besides, it has included some key findings as follows: (1) Given the fact that the universal service fund is not one kind of government funds, shall it still be managed through annual official planning and budgeting? (2) From a viewpoint of profitable undertaking, a fixed network operator might want to charge a remote subscriber outside its service area for some extra facilities fees for service installation. This might be reasonable under commercial terms. But shall we allow such extra fees to be subsidized by the universal service fund? (3) Some people might regard the purpose of the universal service fund as an issue on social security or emergency rescue, and hence distort the spirit of universal service fund. (4) It is questionable whether intangible benefits, resulting form the provision of universal service, could be grouped into revenue forgone. (5) Is it appropriate for an operator to recover its costs of providing coastal radio maritime emergency and safety communication services from the universal service fund? 　　Furthermore, it has given several key policy recommendations as follows: (1) For the sake of dispelling people’s doubts and managing the efficiency in universal service fund, all of the accounts of expenditures shall be open and transparent except that the administrative costs must be kept at minimum. (2) In order to protect consumers’ rights, the equipment installation fee should follow a downward trend and must be cost based. (3) To avoid increasing the burden of telecom operators whose costs are ultimately entailed by consumers, budgets for social security or emergency rescue should be supported by national social welfare policies. (4) Intangible revenue that might be calculable by evaluating average effect of advertisement of each operator or top five operators should be grouped into revenue foregone. Or it may be possible to make a sampling survey of consumers’ behavior, finding the average contribution of each potential customer so as to estimate intangible revenue. (5) Grouping a task force to revise measuring criteria for the facilities and maintenance costs in coastal radio services efficiently. Besides, bids for these services may be opened for volunteers, or be coordinated with and incorporated into the function of Coast Guard Administration, Executive Yuan. (6) It is necessary to build up a mechanism of data management and auditing so as to increase the accountability of re-calculating data. (7) To think over the feasibility that mobile phone can be used for the provision of universal service as well by revising the relevant laws, and then in North, Middle, and South Taiwan, try to conduct a field trial in each area for providing universal service by using mobile phone. (8) For Type I telecommunications businesses, universal service committee can exempt them from universal service obligations with the concerns about their operation efficiency and the duration of losing money, and so on. For Type II telecommunications businesses, only those operating voice communications services, including simple resale and internet phone services, should bear the obligations, but whether they need to make contributions to the universal service fund or not can be decided through the evaluation of their business performance.

本研究所使用違約機率（Probability of Default, 以下簡稱PD）的預測方法為Robust Logistic Regression（穩健羅吉斯迴歸），本研究發展且應用這個方法是基於下列兩個觀察：1. 極端值常常出現在橫剖面資料，而且對於實證結果往往有很大地影響，因而極端值必須要被謹慎處理。2. 當使用Logit Model（羅吉斯模型）估計違約率時，卻忽略極端值。試圖不讓資料中的極端值對估計結果產生重大的影響，進而提升預測的準確性，是本研究使用Logit Model並混合Robust Regression（穩健迴歸）的目的所在，而本研究是第一篇使用Robust Logistic Regression來進行PD預測的研究。 變數的選取上，本研究使用Z-SCORE模型中的變數，此外，在考慮公司的營收品質之下，亦針對公司的應收帳款週轉率而對相關變數做了調整。 本研究使用了一些信用風險模型效力驗證的方法來比較模型預測效力的優劣，本研究的實證結果為：針對樣本內資料，使用Robust Logistic Regression對於整個模型的預測效力的確有提升的效果；當營收品質成為模型變數的考量因素後，能讓模型有較高的預測效力。最後，本研究亦提出了一些重要的未來研究建議，以供後續的研究作為參考。 / The method implemented in PD calculation in this study is “Robust Logistic Regression”. We implement this method based on two reasons: 1. In panel data, outliers usually exist and they may seriously influence the empirical results. 2. In Logistic Model, outliers are not taken into consideration. The main purpose of implementing “Robust Logistic Regression” in this study is: eliminate the effects caused by the outliers in the data and improve the predictive ability. This study is the first study to implement “Robust Logistic Regression” in PD calculation. The same variables as those in Z-SCORE model are selected in this study. Furthermore, the quality of the revenue in a company is also considered. Therefore, we adjust the related variables with the company’s accounts receivable turnover ratio. Some validation methodologies for default risk models are used in this study. The empirical results of this study show that: In accordance with the in-sample data, implementing “Robust Logistic Regression” in PD calculation indeed improves the predictive ability. Besides, using the adjusted variables can also improve the predictive ability. In the end of this study, some important suggestions are given for the subsequent studies.