函數型資料分析為近年發展的統計方法。函數型資料是在一段特定時間上，我們只在離散的時間點上收集觀測值。例如:氣象觀測站所收集到的每月氣溫、雨量資料，即是一種常見的函數型資料。函數型資料主要有三種特色，共同趨勢性、觀測個體反應強度不同，觀測個體時間特色上的差異。本文研究主要是使用，Brumback與Lindstrom在2004所提出的自模型迴歸族(self-modeling)當作模型架構來處理函數型資料的趨勢性與個體反應強度。而為了處理函數型資料的時間差異性，我們在模型中加入時間轉換函數(time transformation function)，處理函數型資料的時間差異性步驟，這個過程稱為同步化。經過同步化的處理後，能幫助研究者更清楚資料的特性。模型中除了時間轉換函數的部份，其餘模型中的參數我們是利用馬可夫鏈蒙地卡羅法中的Gibbs Sampling來進行參數的抽樣，並以取出的抽樣值來估計參數。時間轉換函數的部份，我們使用概似懲罰函數(penalized likelihood function)來估計時間轉換函數的參數部份。由於函數型資料擁有趨勢性，我們預期不同類別的資料，會呈現不同的趨勢性，我們將利用此一特色當做分類上的標準。 關鍵詞:函數型資料分析、曲線同步化、曲線區別分析、馬可夫鏈蒙地卡羅法。 / Functional data are random curves observed in a period of time at discrete time points.They often exhibit a common shape, but with variations in amplitude and phase across curves.To estimate the common shape,some adjustment for synchronization is often made,which is also known as time warping or curve registration.In this thesis,splines are used to model the warping functions and the common shape. Certain parameters are allowed to be random.For the estimation of the random parameters,priors are proposed so that samples from the posteriors can be obtained using Markov chain Monte Carlo methods.For the estimation of non-random parameters, a penalized likelihood approach is used. It is found via simulation studies that for a set of random curves with a common shape,the estimated common shape function looks like the true function up to a location-scale transform,and the curve alignment based on estimated time warping functions looks reasonable.For two groups of random curves which differ in the group common shape functions,synchronization also improves the discrimination between groups in some cases. Key words: functional data analysis,curve registration,curve discrimination,markov chain monte carlo method.

由於違約事件不斷發生以及在財務實證上顯示證券的報酬率有厚尾與高狹峰的現象，本文使用縮減式模型與Lévy過程來評價有信用風險下的可轉換公司債。在Lévy過程中，本研究假設股價服從NIG及VG模型，發現此兩種模型比傳統的GBM模型更符合厚尾現象。此外，在Lévy過程參數估計方面，本文使用最大概似法估計參數，在評價可轉換公司債方面，本研究採用最小平方蒙地卡羅法。本文之實證結果顯示，Lévy模型的績效比傳統GBM模型佳。 / Due to the reason that the default events occurred constantly and still continue taking place, empirical log return distributions exhibit fat tail and excess kurtosis, this paper evaluates convertible bonds under Lévy process with default risk using the reduced-form approach. Under the Lévy process, the underlying stock prices are set to be normal inverse Gaussian (NIG) and variance Gamma (VG) model to capture the jump components. In the empirical analysis, we use the maximum likelihood method to estimate the parameters of Lévy distributions, and apply the least squares Monte Carlo Simulation to price convertible bonds. Five examples are shown in pricing convertible bonds using the traditional model and Lévy model. The empirical results show that the performance of Lévy model is better than the traditional one.

本篇論文的目的在介紹階層貝氏之時間與空間統計模式(spatio-temporal model)，將此模式應用在疾病地圖的分析，以了解疾病在空間上的分佈狀態與時間趨勢。模型中除了納入時間、空間和年齡的效應外，也包括時間與空間、時間與年齡的交互作用，並考慮到空間相關性(spatial correlation)，然後以DIC值(Deviance information criterion)作為模式選取的準則。 本文並以民國88-90年全身紅斑性狼瘡的女性患病人數做為實證分析的資料。配適時間與空間統計模式後，以馬可夫鏈蒙地卡羅法(MCMC)來模擬參數值，估計出各時間、地區、年齡層的對數疾病發生率。由疾病地圖可看出，台灣地區全身紅斑性狼瘡的女性疾病發生率，以20-59歲的年齡層發生率較高，0-19歲的發生率較低。不管在哪一個年齡層，北部和中部地區的發生率都是最高的。時間趨勢方面，88-90年整體疾病發生率有遞減的趨勢，60歲以上的發生率也是遞減的趨勢。但在部分地區，則有發生率遞增的趨勢。 / In this study, we introduce the spatio-temporal model in a hierarchical Bayesian framework and use disease maps to display the spatial patterns and the temporal trends of disease. A special feature of the model is the inclusion of spatial correlations used to examine spatial effects relative to both regional and regional changes over time by group. Then, we use deviance information criterion (DIC) to compare complex hierarchical models. The methodology is illustrated by an analysis of female Systemic Lupls Erythematosus (SLE) morbidity data in Taiwan during the period 1999-2001.The model inference is implemented using Markov chain Monte Carlo method. The outcomes of the practical analysis appear that the higher morbidity rate occurs in 20-year and 40-year period. No matter what age group, the morbidity rate is highest in the north and the middle of Taiwan. Furthermore, the morbidity rate decreases with respect to year as well as over the 60-year period but it increases in some places.

財政部所推動保單紅利自由化及費率自由化政策提供壽險商品市場的另一發展方向，促進我保險市場良性競爭，且自民國九十二年自由分紅保單進入台灣壽險市場，已有數十家壽險公司相繼搶這塊大餅，目前市面上的分紅保單大多採用美式三元利差分紅方式，在紅利部分尚不可設定保證給付，亦即沒有保證保戶每年都一定會領到利差分紅。因此，本研究將介紹在歐洲廣為盛行、附最低保證給付的英式分紅保單，作為國內業者在分紅保單設計上的參考指標。 本研究主要以傳統壽險商品為研究範疇，在不考慮有解約權下以隨機模擬的方式對英式分紅保單進行公平定價。研究方法為在大量模擬下找出不同投資策略的Markowiz效率前緣曲線，並選用這些投資策略於給定不分紅保單的預定利率下，在公平精算原則為前提下以蒙地卡羅法(Monte Carlo Methods)的方式，針對不同目標期末紅利率(Terminal Bonus Target；TB)，找出分紅保單應有的合理預定利率及期間紅利率。再去衡量保險公司不同資產配置下，因為未達到期望紅利所造成的風險。此外，本研究更進一步地以靜態及動態投資策略探討保險公司失去清償能力(Insolvency Risk)的問題。 最後在上述架構下，本研究為了符合市場實際情況，分別針對不同保單期間與不同的繳費方式進行敏感度分析。

衡量投資組合的稀有事件時，即使稀有事件違約的機率極低，但是卻隱含著高額資產違約時所帶來的重大損失，所以我們必須要精準地評估稀有事件的信用風險。本研究係在估計信用損失分配的尾端機率，模擬的模型包含同質模型與異質模型；然而蒙地卡羅法雖然在風險管理的計算上相當實用，但是估計機率極小的尾端機率時模擬不夠穩定，因此為增進模擬的效率，我們利用Glasserman and Li (Management Science, 51(11),2005)提出的重點取樣法，以及根據Chiang et al. (Joural of Derivatives, 15(2),2007)重點取樣法為基礎做延伸的改良式重點取樣法，兩種方法來對不同的投資組合做模擬，更是將改良式重點取樣法推廣至異質模型做討論，本文亦透過變異數縮減效果來衡量兩種方法的模擬效率。數值結果顯示，比起傳統的蒙地卡羅法，此兩種方法皆能達到變異數縮減，其中在同質模型下的改良式重點取樣法有很好的表現，模擬時間相當省時，而異質模型下的重點取樣法也具有良好的估計效率及模擬的穩定性。 / When measuring portfolio credit risk of rare-event, even though its default probabilities are low, it causes significant losses resulting from a large number of default. Therefore, we have to measure portfolio credit risk of rare-event accurately. In particular, our goal is estimating the tail of loss distribution. Models we simulate are including homogeneous models and heterogeneous models. However, Monte Carlo simulation is useful and widely used computational tool in risk management, but it is unstable especially estimating small tail probabilities. Hence, in order to improve the efficiency of simulation, we use importance sampling proposed by Glasserman and Li (Management Science, 51(11),2005) and modified importance sampling based on importance sampling which proposed by Chiang et al. (2007 Joural of Derivatives, 15(2),). Simulate different portfolios by these two of simulations. On top of that, we extend and discuss the modified importance sampling simulation to heterogeneous model. In this article, we measure efficiency of two simulations by variance reduction. Numerical results show that proposed methods are better than Monte Carlo and achieve variance reduction. In homogeneous model, modified importance sampling has excellent efficiency of estimating and saves time. In heterogeneous model, importance sampling also has great efficiency of estimating and stability.

鑒於近年全球海外可轉換公司債發行檔數大增，然而以此商品為研究主題的文獻並不多，於是決定以此為研究目標。 　　影響海外可轉換公司債的價格因素包括股票價格、匯率、國內利率、國外利率和發行公司的違約機率，因此可買回、可賣回海外可轉換公司債是一個複雜的商品，而評價也較為困難。本文採用三維度二項樹和最小平方蒙地卡羅法建立評價海外可轉債的數值模型。為了更貼近真實世界，本文考量各變數間相關性和動態信用風險；另外，為了使評價更為精準，於股價過程中加入跳躍過程。 　　本文將模型運用至兩檔台灣公司所發行的海外可轉債，發現理論價格傾向於高估，但是理論價格與市價極為接近，尤其當以最小平方蒙地卡羅法評價時。另外本文也針對發行條件和模型中各個變數作敏感度分析，其中重要的是發現股票波動度、股票與匯率間相關係數在海外可轉債評價中扮演重要的角色。 / The number of Euro-convertible bonds issued has highly increased in the early 2010s. However, the related literature is barely found. This paper studies the pricing models of this investment product. Euro-convertible bonds are complex instruments affected by the credit risk of the issuers, the dynamic process of stock prices, the term structure of the interest rate and the movement of the exchange rate in the same time. Accordingly, building the ECB pricing model is a hard work. This paper presents a model considering the dynamic credit risk and jump in stock price process to make valuation more precise. Another advantage of models in this paper is use of stochastic interest rates for both local and foreign so as to make the model more staying with the real world. The other advantage is taking the correlation between each random variables into account. For pricing the Euro-convertible bonds, the numerical methodologies used in this paper are three-dimension binomial tree and least squares Monte Carlo approach. For purpose of assessing the performance of the model, two Euro-convertible bonds issued by Taiwan companies are chosen as samples and the difference between the theoretical price and market price during its issue period are provided. The results demonstrate that in spite of pretty slight overestimation, the least squares Monte Carlo simulation does a better job. In addition, this paper performs several kinds of sensitivity analysis to have in-depth understanding about the models. The consequence shows that the volatility of a stock return and the correlation between stock and exchange rate play a central role in ECB valuations.

歐式信用違約交換選擇權通常都以短天期較富流動信，造成這樣情形的原因很可能是因為長天期的信用違約交換選擇權必須承擔標的公司的倒閉風險。美式信用違約交換選擇權讓持有者可以在選擇權到期以前履約，這使得持有者可以只注意信用違約交換溢酬的變動，而不必擔心標的公司的倒閉風險。在這篇論文當中，我們結合最小平方法以及單期信用違約溢酬模型評價美式信用違約交換選擇權，其中單期信用違約溢酬模型是由布瑞格在2004年所發表的模型。本篇論文評價方法的最大優點在於此方法類似於利率理論的市場模型，因此我們可以利用類似的想法評價任何與信用違約交換合約相關的信用衍生性商品。 / The most liquid European CDS options are usually of short maturities. This may result from that options with longer maturity have to bear more default risk of the reference company. American CDS options allow the holders to exercise options before option matures so that they can focus on spread movements without worrying about default risk. In this paper, we price American CDS options with one-period CDS spread model presented by Brigo (2004). The primary advantage of this model is that it is similar to LIBOR market model in interest rate theory. Therefore, path-dependent CDS-related products can be easily priced with familiar ideas.

本文針對躉繳保費的附保證變額壽險進行評價，保單形式為生死合險，假設投保人可將期初的投資金額連結到兩種投資標的：股價指數及債券型基金，並以BGM模型描述利率的動態過程，然後分別計算不含解約權及含解約權的附保證變額壽險躉繳保費，進而求算出隱含在保單中的保證價值和解約權價值。針對含解約權的附保證變額壽險，以Longstaff and Schwartz(2001)提出的最小平方蒙地卡羅法處理解約的問題。最後，我們求算不同年齡下的男性保費，並且在投資比例、起始最低保證、最低保證給付成長率、針對解約的保證給付成長率和第一個允許的解約時點變動下，分別討論對於保證價值和解約權價值的影響。 結果顯示：(1)當起始最低保證給付等於期初投資金額時，投資在股票的比例越大，越能凸顯保證價值和解約權價值佔保費的比重。以30歲男性為例，保證價值佔不含解約權之附保證變額壽險的比例，由全部投資在債券型基金的0.03%，成長到全部投資在股票的13.86%；而解約權價值佔含解約權之附保證變額壽險的比例，由全部投資在債券型基金的0.05%，成長到全部投資在股票的9.12%。(2)投資比例、起始最低保證給付和最低保證給付成長率越大，保證價值越高。(3)起始最低保證給付和針對解約的保證給付成長率越大，解約權價值越大；而最低保證給付成長率和第一個允許的解約時點越大，解約權價值越小。(4)投資比例隨著最低保證給付不同對解約權價值有不同的影響。 關鍵字：附保證變額壽險、BGM利率模型、解約選擇權、最小平方蒙地卡羅法 / This study emphasizes on the pricing of variable life insurance with minimum guarantees. As an endowment policy in a single premium form, in this paper, it is assumed that the insured can distribute the initial investment amount into two underlying assets: the stock index fund and bond fund. Simulating the interest rate under a BGM model, computational procedures are performed for the single premium of the variable life insurance policy without surrender option and embedding a surrender option, and further, the guarantee value and surrender value embedded in the insurance policy. For the variable life insurance policy embedding a surrender option, the Least Square Monte-Carlo method proposed by Longstaff and Schwartz (2001) is applied to solve the surrender conditions. Finally, we calculate the premium for a male at different ages, and respectively analyze the variations of the guarantee value and surrender value under the influence of the investment portfolio, the initial minimum guaranteed amount, the growth rate of the minimum guarantee, the growth rate of the minimum guarantee for surrender and the first permitted surrender time. The results show that: (1) when the initial minimum guaranteed amount equals the initial investment amount, higher proportion invested in stock will result in larger percentage of the guarantee value and surrender value to total premium. Take a 30-year old male as an example: the percentage of guarantee value to the premium of variable life insurance with minimum guarantee and without a surrender option, which is 0.03% when the initial investment amount thoroughly goes to bond fund, rises up to 13.86% with the entire amount invested in stock index fund. Likewise, the percentage of surrender value to the premium of variable life insurance with minimum guarantee and surrender option is 0.05% with total amount invested in bond fund, while it is 9.12% with the entire amount invested in stock index fund. (2) The higher proportion invested in stock, the initial minimum guaranteed amount and the growth rate of minimum guaranteed amount, the larger guarantee value. (3) Larger initial minimum guaranteed amount and the growth rate of the minimum guaranteed amount for surrender would contribute to a higher surrender value. The higher growth rate of the minimum guaranteed amount and the first permitted surrender time, the lower surrender value. (4) The influence of the investment portfolio to surrender value depends on the initial minimum guaranteed amount. Key words: Variable life insurance with minimum guaranteed amount, BGM interest rate model, surrender option, least squares Monte Carlo approach.

本論文主要在估計門檻式自動迴歸模型之參數的信賴區間。由線性自動迴歸 模型衍生出來的非線性自動迴歸模型中，門檻式自動迴歸模型是其中一種經常會被應用到的模型。雖然，門檻式自動迴歸模型之參數的漸近理論已經發展了許多；但是，相較於大樣本理論，有限樣本下參數的性質討論則較少。對於有限樣本的研究，Woodroofe (1989) 提出一種近似法：非常弱近似法。 Woodroofe 和 Coad (1997) 則利用此方法去架構一適性化線性模型之參數的修正信賴區間。Weng 和 Woodroofe (2006) 則將此近似法應用於線性自動迴歸模型。這個方法的應用始於定義一近似樞紐量，接著利用此方法找出近似樞紐量的近似期望值及近似變異數，並對此近似樞紐量標準化，則標準化後的樞紐量將近似於標準常態分配，因此得以架構參數的修正信賴區間。而在線性自動迴歸模型下，利用非常弱展開所導出的近似期望值及近似變異數僅會與一階動差及二階動差的微分有關。因此，本論文的研究目的就是在樣本數為適當的情況下，將線性自動迴歸模型的結果運用於門檻式自動迴歸模型。由於大部分門檻式自動迴歸模型的動差並無明確之形式；因此，本研究採用蒙地卡羅法及插分法去近似其動差及微分。最後，以第一階門檻式自動迴歸模型去配適美國的國內生產總值資料。 / Threshold autoregressive (TAR) models are popular nonlinear extension of the linear autoregressive (AR) models. Though many have developed the asymptotic theory for parameter estimates in the TAR models, there have been less studies about the finite sample properties. Woodroofe (1989) and Woodroofe and Coad (1997) developed a very weak approximation and used it to construct corrected confidence sets for parameters in an adaptive linear model. This approximation was further developed by Woodroofe and Coad (1999) and Weng and Woodroofe (2006), who derived the corrected confidence sets for parameters in the AR(p) models and other adaptive models. This approach starts with an approximate pivot, and employs the very weak expansions to determine the mean and variance corrections of the pivot. Then, the renormalized pivot is used to form corrected confidence sets. The correction terms have simple forms, and for AR(p) models it involves only the first two moments of the process and the derivatives of these moments. However, for TAR models the analytic forms for moments are known only in some cases when the autoregression function has special structures. The goal of this research is to extend the very weak method to the TAR models to form corrected confidence sets when sample size is moderate. We propose using the difference quotient method and Monte Carlo simulations to approximate the derivatives. Some simulation studies are provided to assess the accuracy of the method. Then, we apply the approach to a real U.S. GDP data.

針對金融時序資料變異數不齊一的性質，隨機波動模型除了提供於ARCH族外的另一選擇；且由於其設定隱含波動本身亦為一個隨機波動函數，藉由設定隨時間改變且自我相關的條件變異數，使得隨機波動模型較ARCH族來得有彈性且符合實際。傳統上處理隨機波動模型的參數估計往往需要面對到複雜的多維積分，此問題可藉由貝氏分析裡的馬可夫鏈蒙地卡羅法解決。本文主要的探討標的，即在於利用馬可夫鏈蒙地卡羅法估計美元/新台幣匯率隨機波動模型參數。除原始模型之外，模型的擴充分為三部分：其一為隱含波動的二階自我回歸模型；其二則為藉由基本模型的修改，檢測匯率市場上的槓桿效果；最後，我們嘗試藉由加入scale mixture的方式以驗證金融時序資料中常見的厚尾分配。