第一章：導論。分三節。敘述本論文之研究目的、研究假設；研究架構；以及名詞操 作性定義。 第二章：理論及文獻探討。共四節。包括價值理論；生活型態理論；價值觀與生活型 態理論；以及其他相關理論之研究與文獻探討。 第三章：研究方法。共四節。分別敘述本論文之研究對象、抽樣方法、研究工具以及 分析方法。 第四章：研究結果與討論。分二節。討論變項分析結果以及假設驗證結果。 第五章：結論。共三節。分別敘述研究結論、研究限制、建議事項。

第一章為緒論, 將偏誤的影響作一討論。第二章為無反應誤差, 研究其效應並利用訪 查法以減少誤差。第三章為測量誤差, 先建立模型, 將均方誤分解, 再討論測量誤差 對於均方誤各分量之影響, 最後為測量誤差之研究方法—貫穿子樣本法及比較法。第 四章為敏感問題, 敏感問題甚易造成非抽樣誤差, 故建立隨機反應法及無關問題法 , 以推定母體之比例或平均數。第五章為實證分析, 將第四章的方法應用於實際問題— 求台北市內兒科醫師每月平均收入, 並和傳統直接問題法比較分析, 最後作檢討與改 進。第六章為結論, 將上述各種方法予以綜合說明並加以補充。

In the wake of growing database that has already become the trend of today’s business environment within the foreseeable future, reviewing quality information from mountains of data residing on corporations or organizations’ network such as sales figures, manufacturing statistics, financial data and experimental data is clearly costly, time consuming and definitely ineffective approach. Therefore we would need a sound and effective method in obtaining only portions of the data that are representative to the population and which allow us to build the reliable model based upon the sampled data. However, sometimes we have a situation where the database is of limited in size, under such circumstance, we initiate the idea which is relatively new to adding the attributes or values into the database to enhance the quality of the data Follow through such a procedure; it is obvious that implementing a good sampling method is an important groundwork leading us to reach final destination that is obtaining a reliable predictive model. And this is our research goal that is to get an effective and representative value-added sample of by means of sampling method for building an accuracy predictive model. The concept is pretty straightforward that is if we want to get good predictive samples then we need the correct sampling methods. The sampling methods under study are simple random sample, system sample, stratified sample and uniform design. The models used are the C5.0, logistic regression, and neural network for categorical predictive variable and stepwise regression for continuous predictive variable. The results are discussed in the conclusion section. Keywords: Database、Data Mining、Sampling、Value-added database

The over-adjustment means that the process is adjusted unnecessarily when a false alarm occurs. It may result in shifts in process mean and variance affecting the quality of products and have the effect of an increase in variability and cost. In this paper, the economic variable sample interval (VSI) standard and control charts are proposed to monitor effectively the mean and variance of the over-adjusted process. We use a Markov chain approach to derive the design parameters of the standard and charts by the minimum of the cost function. An example of shampoo making process is used to illustrate the application and performance of the proposed VSI standard and charts in detecting shifts in process mean and variance. Furthermore, we compare the cost and performances for the economic FSI (fixed sampling interval) and VSI control charts. Support for this research was provided in part by the National Science Council of the Republic of China, grant No. NSC 94-2118-M-004-003. / The over-adjustment means that the process is adjusted unnecessarily when a false alarm occurs. It may result in shifts in process mean and variance affecting the quality of products and have the effect of an increase in variability and cost. In this paper, the economic variable sample interval (VSI) standard and control charts are proposed to monitor effectively the mean and variance of the over-adjusted process. We use a Markov chain approach to derive the design parameters of the standard and charts by the minimum of the cost function. An example of shampoo making process is used to illustrate the application and performance of the proposed VSI standard and charts in detecting shifts in process mean and variance. Furthermore, we compare the cost and performances for the economic FSI (fixed sampling interval) and VSI control charts. Support for this research was provided in part by the National Science Council of the Republic of China, grant No. NSC 94-2118-M-004-003.

金融資產報酬的波動一直都是財務市場熱衷研究的主題, 由於真正報酬的波動無法確知, 造成無法判斷何者為衡量報酬波動最佳的模型, 進而導致預測未來報酬的風險增加。因此, 本文利用靴帶抽樣法(Bootstrap)反覆抽樣的估計方式, 建立報酬與報酬波動的預測區間來衡量由估計模型參數產生的不確定性, 希望能藉此更瞭解資產報酬的變化以降低投資風險。鑒於目前衡量報酬波動的模型眾多, 文中將採用文獻上普遍最能掌握金融資產報酬波動現象的GARCH模型, 作為衡量報酬波動的方法, 再以靴帶抽樣方法估計其報酬與報酬波動的預測區間, 透過有限樣本的模擬將估計模型參數不確定性的靴帶抽樣方法與其他方法比較, 證明靴帶抽樣法最能適當的捕捉報酬波動真實的情況。最後, 由台灣上市股票市場中選取四支不同類股的各股以日報酬進行實證研究, 結果顯示各股的日報酬都具有波動變異的現象, 進一步估計樣本外不同範圍的波動預測區間, 發現利用估計模型參數不確定性的靴帶抽樣方法可以適當地涵蓋波動的變化。

疾病就診率係針對固定期問內，患有疾病而到全民健康保險特約之各級醫療院所就診之人數予以統計，其分析可供規劃醫療保健服務與修訂全民健保制度之參據，因此應成為經常性之統計工作。本文以適合連續性調查的輪換抽樣設計為原則，運用調查當期及較早期之資訊以建搆複合估計式，並推導該估計式之期望值與變異數，據以此比較與簡單估計式之相對效率。為評估複合估計方法運用於疾病就診率之效益，遂選定某些疾病進行分析，並研究在何種權數組合下，複合估計式能減低變動情形及其相對效率。 關鍵字：疾病就診率、輪換抽樣法、複合估計式、相對效率 / The treatment rate is a statistical analysis of numbers of people who had taken treatment in the fixed period. It can be a reference of planning medical service, revising medical laws and so on. So its analysis is important and should be an occasionally statistical work. In this paper, we adopt rotation sampling method, combine current and earlier information to construct composite estimator, and derive its expected value and variance formula, so as to compare with simple estimator. To evaluate the benefit of adopting composite estimator on the treatment rate, we choose some diseases as objectives, and determine the optimal weighted parameter so that composite estimator can reduce the variance and its effects on the relative efficiency. Key words: treatment rate, rotation sampling, composite estimator, relative efficiency

本研究以某國中之某一班級學生為研究對象，藉由長時間觀察班級友誼動態網絡圖之變化，來了解班級內同學之間的互動和交友情形。 在研究方法上，兼採量化與質化研究。量化部分，本研究採用自製問卷，自2010年9月至2011年10月，共計進行九次問卷，將問卷以NETDRAW軟體製成動態網絡圖，並以UCINET軟體做資料分析及密度檢定，比較各次問卷之間的友誼網絡密度，是否會受到學校內活動或其他特殊事件的影響；質化部分，配合量化分析結果，以教師觀察、訪談、校園活動事件觀點，分析友誼網絡變化的可能原因，期能成為教師在班級團體輔導上的參考。 根據研究發現，本研究的結論如下： 一、班級內友誼動態網絡圖可幫助導師掌握同學交友情形。 二、學生的友誼網絡及網絡密度可能會受到校內活動或特殊事件而改變。 三、班級內的同儕團體中，男生可分為大團體、小團體，而女生可分為活躍主導團體、非主流團體、內向團體和邊際團體。 四、班級內同儕團體間的互動情形是：女生的界線明顯，互動不多；男生界線不明顯，互動頻繁。 五、男女生的友誼網絡密度不同，其中原因可能是對「朋友」定義的認知差異。 六、導師對於班級內受排擠同學，若未及時處理，則不易立即改變被排擠的現象。 綜合以上，本研究提出的建議如下： 一、未來的研究者，可多以青少年為研究對象，並以量化和質化研究並用之方式進行。 二、導師應即時給予被孤立者協助。 三、可透過不同的分組策略，增加班級內彼此不熟悉同學的互動。 四、導師可透過問卷，掌握學生的交友狀況。 五、學校可以開設交友相關課程，讓青少年懂得如何與人相處。

有賴電腦的輔助，企業或組織內部所存放的資料量愈來愈多，加速資料量擴大的速度。但是大量的資料帶來的未必是大量的知識，即使擁有功能強大的資料庫系統，倘若不對資料作有意義的分析與推論，再大的資料庫也只是存放資料的空間。過去企業或組織只把資料庫當作查詢系統，並不知道可以藉由資料庫獲取有價值的資訊，而其中資料庫的內容完整與否更是重要。由於企業所擁有的資料庫未必健全，雖然擁有龐大資料庫，但是其中資訊未必足夠。我們認為利用資料庫加值方法：插補方法、抽樣方法、模型評估等步驟，以達到擴充資訊的目的，應該可以在不改變原始資料結構之下增加資料庫訊息。 本研究主要在比較不同階段的資料經過加值動作後，是否還能與原始資料結構一致。研究架構大致分成三個主要流程，包括迴歸模型、羅吉斯迴歸模型與決策樹C5.0。經過不同階段的資料加值後，我們所獲得的結論為在迴歸模型為主要流程之下，利用迴歸為主的插補方法可以使加值後的資料庫較貼近原始資料，若想進一步採用抽樣方法縮減資料量，系統抽樣所獲得的結果會比利用簡單隨機抽樣來的好。而在決策樹C5.0的主要流程下，以類神經演算法作為插補的主要方法，在提增資訊量的同時，也使插補後的資料更接近原始資料。關於羅吉斯迴歸模型，由於間斷型變數的類別比例差異過大，致使此流程無法達到有效結論。 經由實證分析可以瞭解不同的配模方式，表現較佳的資料庫加值技術也不盡相同，但是與未插補的資料庫相比較，利用資料庫加值技術的確可以增加資訊量，使加值後的虛擬資料庫更貼近原始資料結構。 / With the fast pace of advancement in computer technology, computers have the capacity to store huge amount of data. The abundance of the data, without its proper treatment, does not necessary mean having valuable information on hand. As such, a large database system can merely serve as ways of accessing and storing. Keeping this in mind, we would like to focus on the integrity of the database. We adapt the methods where the missing values are imputed and added while leaving the data structure unmodified. The interest of this paper is to find out when the data are post value added using three different imputation methods, namely regression analysis, logistic regression analysis and C5.0 decision tree, which of the methods could provide the most consistent and resemblance value-added database to the original one. The results this paper has obtained are as the followings. The regression method, after imputation of the added value, produced the closer database structure to the original one. And in the case of having large amount of data where the smaller size of data is desired, then the systematic sampling provides a better outcome than the simple random sampling. The C5.0 decision tree method provides similar result as with the regression method. Finally with respect to the logistic regression analysis, the ratio of each class in the discrete variables is out of proportion, thereby making it difficult to make a reasonable conclusion. After going through the above studies, we have found that although the results from three different methods give slight different outcomes, one thing stands out and that is using the technique of value-added database could actually improve the authentic of the original database.

傳統的保險人在面對保險契約所承保的風險時，常會藉由國際上的再保險市場來分散其保險風險。由於所承保險事件的不確定性，保險人需要謹慎小心評估其保險風險並將承保風險轉移至再保險人。再保險有兩種主要的保險型式，可區分成比例再保契約及超額損失再保契約，保險人將利用這些再保險契約來分散求償給付時的損失，加強保險人本身的財務清償能力。 本研究，主要在於建構未來損失求償幅度或頻率的預測分佈並模擬未來支付求償的損失。簡單重點重複抽樣法是一種從危險參數的驗後分佈中抽樣的抽樣方法。然而，蒙地卡羅模擬是一種利用大量電腦運算計算近似預測分佈的逼近方法。利用被選取危險參數的驗前分佈來模擬其驗後分佈，並建構可能的承保危險參數結構，將基於馬可夫鏈蒙地卡羅理論的吉普生抽樣方法決定最適自留額，同時運用於再保險合約決策擬定過程。 最後，考慮於不同的再保險契約下來衡量再保險人的自負財務風險。基本上我們研究的對象是針對保險人所承保的風險，再藉由上述的方法來模擬、近似以量化所衍生的財務風險。這將有助於保險人清楚地瞭解其承保的風險，並對其承保業務做妥善的財務風險管理。本研究提供保險人具體的模型建構方法並對此建構技巧做詳細說明及實證分析。 / Insurers traditionally transfer their insurance risk through the international reinsurance market. Due to the uncertainty of these insured risks, the primary insurer need to carefully evaluate the insured risk and further transfer these risks to his ceding reinsurers. There are two major types of reinsurance, i.e. pro rata treaty and excess of loss treaty, used in protecting the claim losses. In this article, the predictive distribution of the claim size is constructed to monitor the future claim underwriting losses based on the reinsurance agreement. Simple Importance Resampling (SIR) are employed in sampling the posterior distribution of risk parameters. Then Monte Carlo simulations are used to approximate the predictive distribution. Plausible prior distributions of these risk parameters are chosen in simulation its posterior distribution. Markov chain Monte Carlo (MCMC) method using Gibbs sampling scheme is also performed based on possible parametric structures. Both the pro rata and excess of loss treaties are investigated to quantify the retention risks of the ceding reinsurers. The insurance risks are focused in our model. Through the implemented model and simulation techniques, it is beneficial for the primary insurer in projecting his underwriting risks. The results show a significant advantage and flexibility using this approach in risk management. This article outlines the procedure of building the model. Finally a practical case study is performed for numerical illustrated.

每個人在面臨特定的選擇集合時, 可能會有不同的選擇行為發生, 如果將每個人的反 應集合起來, 組成一組資料, 利用這些資料, 我們就可以導出一種行為模式。一般的 行為模式都將應變數視為連續變數( 例如需求量),近年發展出的個體選擇模式(Discr -ete Choice Model)則將應變數視為間斷的(Discrete)孌數( 例如交通工具的選擇: 汽車=1、火車=2、飛機=3等等) 。藉由這種模式, 我們可以進一步研究或預測人類在 面臨各種選擇時所產生的行為。 一個行為模式, 包含了可以被觀察的自變數和未知的參數, 我們利用抽樣觀察而得的 資料來推估這些未知的參數, 但不同的抽樣方法會導致不同的估計值。一般而言, 在 這類問題中應用到的抽樣方法有三種:1. 簡單隨機抽樣、2.外生分層抽樣、3.內生分 層抽樣。本文主要的研究內容集中在探討運用不同的內生分層抽樣法所推估的參數值 之間的差異。 研究方法大致可分為三個步驟: 一. 閱讀前人的文獻, 彙總整理前人的方法, 并發現新的抽樣與估計法: 此法是先決 定內生分層后每層的抽樣比例, 再用系統抽樣的方法抽出樣本, 最后用最大概似法對 參數加以推估。以Binary Logit Model為例: P(1︱x)=1/(1+exp(-xb)) P(0︱x)=1/(1+exp(xb)) 其log likelihood fuction為 logL=Σ {y [log(gl)+log(p(1︱x )]+(1-y )[log(g0)+log(P(0︱x )]-log[gl P (1︱x )+g0 P(0︱x )]} 其中y =1 or 0;gi是選擇i 選項的抽樣比例(i=1 or 0);x 為自變數;b是欲估計的參 數。 二、證明利用新法所產生之估計式, 和前人所研究出的估計式同樣具有一致性。 三、模擬一組母體, 再比較新法與前人的估計式間的差異。