Hamilton (1989)發展出馬可夫轉換模型，假設模型母體參數會隨某一無法觀察得到的狀態變數變動而改變，並用馬可夫鏈的機制來掌控狀態間切換，可適當掌握金融與經濟變數所面臨的結構改變，因此是一個十分重要的財務模型。Schwert (1989)觀察股價波動狀況，發現經濟衰退期的股價波動比經濟擴張期大，因此認為Hamilton (1989)所提出的馬可夫轉換模型亦可應用於股票市場。然而，發現當市場上有重大訊息來臨時，大部分標的資產報酬率會產生跳躍現象，因此汪昱頡 (2008)提出跳躍風險下馬可夫轉換模型，以改善馬可夫模型所無法反映之股價不正常跳躍現象。在探討股價指數報酬率之敘述統計量與動態圖後，本文認為跳躍幅度也會受狀態影響，因此進一步拓展周家伃 (2010)跳躍獨立風險下狀態轉換模型，期望對股市報酬率動態過程提供更佳的分析。實證部分使用1999到2010年的國際股價指數之S&P500、道瓊工業指數與日經225三檔作為研究資料，來說明股價指數具有狀態轉換及跳躍的現象，並利用EM(Expectation Maximization)演算法來估計模型的參數，以SEM(Supplemented Expectation Maximization )演算法估計參數的標準差，且使用概似比(Likelihood ratio)檢定結果顯示跳躍相關風險下狀態轉換模型比跳躍獨立風險下狀態轉換模型更適合描述股價指數報酬率。最後，驗證跳躍相關風險下狀態轉換模型能捕捉其報酬率不對稱、高狹峰與波動聚集之特性。 / Hamilton (1989) proposed Markov switching models to suppose the model parameters change with unobserved state variables which control the switch between states by Markov chain. It can be appropriate to grasp the financial and economic variables which facing structural changes, so it’s a very important financial model. Schwert (1989) observed stock prices, and discovered that the volatilities of recession are higher than the volatilities of expansion. Hence, Schwert (1989) suggested to apply the Markov switching models to stock market. However, most of underlying asset return have jump phenomenon when abnormal events occur to financial market. Wong (2008) proposed Markov switching models with jump risks to improve Markov switching models which can not capture the jump risk of asset price. According to stock index return’s descriptive statistics and dynamic graph, we argue that states will impact jump sizes. In this paper, we extend the regime-switching model with independent jump risks (Chou, 2010) to provide better analysis for the dynamic of return. This paper use stock indices of the study period from 1999 to 2010 to estimate the parameters of the model and variance of parameter estimators by Expectation-Maximization (EM) algorithm and SEM(Supplemented Expectation Maximization ) , respectively. And use the likelihood ratio statistics to test which model is appropriate.Finally, the empirical results show that regime-switching model with jump sizes dependency risk can capture leptokurtic feature of the asset return distribution and volatility clustering phenomenon.

因為疾病空間模式通常會與環境中的危險因子有很強烈的關聯性，因此流行病學家與社會大眾都對疾病的空間模式感到興趣。舉例來說，空間群聚就是一項非常受到重視的疾病空間模式，在眾多的空間群聚檢定方法種，Kulldorff和 Nagarwalla在1995年提出的空間掃描統計量是相當受到廣泛應用的方法，雖然這個統計方法可以檢定初空間資料的異質性，但是卻沒有辦法區隔這些異質性是來自於整體空間資料的相關性或是局部的空間群聚。在本篇論文中，我們將分別提出計次型的統計方法與貝氏統計方法兩種類型的空間群聚檢定方法來處理這樣的問題，其中計次型的統計方法為一兩階段的統計方法，首先採用EM演算法來估計空間自相關，並根據估計的結果與掃描窗格在偵測空間群聚；另一方面，貝氏方法則考慮加入群聚的中心位置及半徑作為事前的機率分布，進而透過MCMC的方法來計算出後驗分布的結果。除此之外，北卡羅來納的嬰兒猝死症和台灣老年人口癌症死亡資料將被用來示範與評價不同群聚檢定方法的差異與效果。

銀行在評分模型建置完成後需進行驗證工作，以瞭解評分模型是否能有效評出客戶的風險層級，穩健地估計區別鑑別力指標為驗證工作中的重點。在先前的文獻中假設正常授信戶與違約戶分數分配為常態分配。但在實際資料中，分配未必定為常態。因此本文接著探討在正常授信戶與違約授信戶之分配為混合分配，即兩分數分配為偏斜常態分配下，何種方法可以對於估計AUC具有較高的穩定性。本文比較五種估計AUC的方法，分別為常態核，經驗分配，曼惠尼近似，最大摡似法和EM演算法。模擬結果呈現(1)投信戶組合分配為兩常態分配下，最大摡似法在大部分違約率下都可以得到較窄的信賴區間。(2)組合分配為一常態與一偏斜常態及兩偏斜常態分配下，EM演算法在大部分情況有較窄的信賴區間，其中在兩偏斜常態分配下，表現更佳。(3)曼惠尼近似建構的信賴區間寬度最大，代表曼惠尼近似是較保守的估計方法。 / Banks face discrimination after constructing the rating systems to figure out whether the systems can discriminate defaulting and non-defaulting borrowers. Literature assumed the two score distribuion are normal distributed. However, the real data may not be normal distribuions. We assum the two score distribuions are skewed normal distribuions to discuss which method has more robustness to estimate the AUC value.Under skewed distribution, we propose EM algorithm to estimate the population parametric. If used properly, information about the population properties may be used to get better accuracy of estimation the AUC value.Numerical results show the EM algorithm method , comparing with other methods, has robustness in detect the rating systems have discirmatory power.

本文研究美國退休福利保險公司(PBGC)保險價值的計算,延伸 Marcus (1987)模型,提出狀態轉換過程保險價值模型計算,也就是將市場分為兩種情況,正成長率視為正常狀態,負成長率為衰退狀態,利用狀態轉換過程評價 PBGC 契約在經濟困難而終止和介入終止下合理的保險價值。在參數估計方面,本文以 S&P500股價指數和一年期國庫券資料參數估計值及Marcus(1987)和Pennacchi and Lewis(1994)的方式給定參數,以 EM-PSO-Gradient 延伸 EM-Gradient 方法並以最大概似函數值、AIC 準則和 BIC 準則比較估計結果。最後固定其他參數, 探討狀態轉換過程保險價值模型對參數調整後保險價值的影響之敏感度分析。 / In this paper, we evaluate Pension Benefit Guaranty Corporation insurance values through regime switching models, which is the extension of the models of Marcus (1987). That is, we can separate periods of economy with faster growth from those with slower growth when observing long-term trends in economy and calculate the reasonable PBGC insurance values under distress termination and intervention termination by regime switching processes. We set parameters by estimating S&P 500 index and 1-year treasury bills by EM-PSO-Gradient, which is the extensive method of EM-Gradient and refer the methods of setting parameters from Marcus (1987) and Pennacchi and Lewis (1994). After that, we compare the maximum likelihood estimates, AIC and BIC of the estimative results. Finally, we do sensitivity analysis through given the other parameters and look into what would impact on our models of insurance values when adjusting one parameter.

我們提供一個新的可靠度模型，DwACM，並提供一個模式選擇準則CCP，我們利用DwACM和CCP來選擇衰變量。 / We propose a new reliability model, DwACM (Dichotomous-data with Auxiliary Continuous Measurements model) to describe a data set which consists of classical dichotomous response (Go or No Go) associated with a set of continuous auxiliary measurement. In this model, the lifetime of each individual is considered as a latent variable. Given the value of the latent variable, the dichotomous response is either 0 or 1 depending on if it fails or not at the measuring time. The continuous measurement can be regarded as observations of an underlying possible degradation candidate of which descending process is a function of the lifetime. Under the assumption that the failure of products is defined as the time at which the continuous measurement reaches a threshold, these two measurements can be linked in the proposed model. Statistical inference under this model are both in frequentist and Bayesian frameworks. To evaluate the continuous measurements, we provide a criterion, CCP (correct classification probability), to select the best degradation measurement. We also report our simulation studies of the performances of parameters estimators and CCP.

本文主要著重在處理條件評估法下，「不知道」受訪者的回應。當「不知道」受訪者的產生機制並未符合完全隨機時，考量他們的真實意向就顯得極為重要。 文中使用中央研究院生醫所在其研究計畫「竹東及朴子地區心臟血管疾病長期追蹤研究」（CardioVascular Disease risk FACtor Two-township Study，簡稱CVDFACTS）第五循環中的研究調查資料。 　　由於以往的文獻對於「不知道」受訪者的處理，皆有不足之處。如Wang (1997)所提出的方法，就只能針對某種特定的「不知道」受訪者來做處理；而Caudill and Groothuis (2005)所提的方法，由於將「不知道」受訪者的差補與願付價格的估計分開，亦使其估計結果不具備一些好的性質。在本文中，我們提出一個能同時處理「不知道」受訪者且估計願付價格的方法。除了使得統計上較有效率外，也保有EM演算法的一個特性：願付價格模型中的估計參數為最大概似估計值。此外，在加入三要素混合模型(Tsai (2005))後，我們也可避免用到極端受訪者的訊息去差補那些「不知道」受訪者的意向。 　　在分析願付價格的過程中，我們發現此筆資料的「不知道」受訪者，其產生的機制為隨機，而非為完全隨機，這意謂著不考量「不知道」受訪者的分析結果，必定會產生偏差。而在比較有考量「不知道」受訪者與沒有的情況後，其結果確實應證了我們的想法：只要「不知道」受訪者不是完全隨機產生的，那麼不考量他們必定會產生某種程度的偏差。 / This paper investigates how to deal with “Don’t Know” (DK) responses in contingent valuation surveys, which must be taken into consideration when they are not completely at random. The data we use is collected from the fifth cycle of the Cardiovascular Disease Risk Factor Two-township Study (CVDFACTS), which is a series of long-term surveys conducted by the Institute of Biomedical Sciences, Academia Sinica. Previous methods used in dealing with DK responses have not been satisfactory because they only focus on some types of DK respondents (Wang (1997)), or separate the imputation of DK responses from the WTP estimation (Caudill and Groothuis (2005)). However, in this paper, we introduce an integrated method to cope with the incomplete data caused by DK responses. Besides being more efficient, the single-step method guarantees maximum likelihood estimates of the WTP model to be obtained due to the good property that the EM algorithm possesses. Furthermore, by adding the concept of the three-component mixture model (Tsai (2005)), some extreme information are drawn out when imputing the DK inclinations. In this hypertension data, the mechanism of the DK responses is “Don’t know at random”, which means the analysis of DK-dropped results in a bias. By using our method, the difference between DK-dropped and DK-included is actually revealed, which proves our suspicion that a DK-dropped analysis is accompanied by a biased result when DK is not completely at random.

在處理具遺失值之類別資料時，傳統的方法是將資料捨棄，但是這通常不是明智之舉，這些遺失某些分類訊息的資料通常還是可以提供其它重要的訊息，尤其當這類型資料的個數佔大多數時，將其捨棄可能使得估計的變異數增加，甚至影響最後的決策。如何將這些遺失某些訊息的資料納入考慮，作出完整的分析是最近幾十年間頗為重要的課題。本文主要整理了五種分析這類型資料的方法，分別為單樣本方法、多樣本方法、概似方程式因式分解法、EM演算法，以上四種方法可使用在資料遺失呈隨機分佈的條件成立下來進行分析。第五種則為樣本遺失不呈隨機分佈之分析方法。 / Traditionally, the simple way to deal with observations for which some of the variables are missing so that they cannot cross-classified into a contingency table simply excludes them from any analysis. However, it is generally agreed that such a practice would usually affect both the accuracy and the precision of the results. The purpose of the study is to bring together some of the sound alternatives available in the literature, and provide a comprehensive review. Four methods for handling data missing at random are discussed, they are single-sample method, multiple-sample method, factorization of the likelihood method, and EM algorithm. In addition, one way of handling data missing not at random is also reviewed.

Black and Scholes (1973)對於報酬率提出以B-S模型配適，但B-S模型無法有效解釋報酬率不對稱高狹峰、波動度微笑、波動度叢聚、長記憶性的性質。Merton (1976)認為不尋常的訊息來臨會影響股價不連續跳躍，因此發展B-S模型加入不連續跳躍風險項的跳躍擴散模型，該模型可同時描述報酬率不對稱高狹峰和波動度微笑兩性質。Charles, Fuh and Lin (2011)加以考慮市場狀態提出狀態轉換跳躍模型，除了保留跳躍擴散模型可描述報酬率不對稱高狹峰和波動度微笑，更可以敘述報酬率的波動度叢聚和長記憶性。本文進一步拓展狀態轉換跳躍模型，考慮不連續跳躍風險項的帄均數與市場狀態相關，提出狀態轉換跳躍相關模型。並以道瓊工業指數與S&P 500指數1999年至2010年股價指數資料，採用EM和SEM分別估計參數與估計參數共變異數矩陣。使用概似比檢定結果顯示狀態轉換跳躍相關模型比狀態轉換跳躍獨立模型更適合描述股價指數報酬率。並驗證狀態轉換跳躍相關模型也可同時描述報酬率不對稱高狹峰、波動度微笑、波動度叢聚、長記憶性。最後利用Esscher轉換法計算股價指數選擇權定價公式，以敏感度分析模型參數對於定價結果的影響，並且市場驗證顯示狀態轉換跳躍相關模型會有最小的定價誤差。 / Black and Scholes (1973) proposed B-S model to fit asset return, but B-S model can’t effectively explain some asset return properties, such as leptokurtic, volatility smile, volatility clustering and long memory. Merton (1976) develop jump diffusion model (JDM) that consider abnormal information of market will affect the stock price, and this model can explain leptokurtic and volatility smile of asset return at the same time. Charles, Fuh and Lin (2011) extended the JDM and proposed regime-switching jump independent model (RSJIM) that consider jump rate is related to market states. RSJIM not only retains JDM properties but describes volatility clustering and long memory. In this paper, we extend RSJIM to regime-switching jump dependent model (RSJDM) which consider jump size and jump rate are both related to market states. We use EM and SEM algorithm to estimate parameters and covariance matrix, and use LR test to compare RSJIM and RSJDM. By using 1999 to 2010 Dow-Jones industrial average index and S&P 500 index as empirical evidence, RSJDM can explain index return properties said before. Finally, we calculate index option price formulation by Esscher transformation and do sensitivity analysis and market validation which give the smallest error of option prices by RSJDM.

退休金是退休人未來生活的依靠，確保在退休後能得到適足的退休給付，政府在退休金上實施保證收益制度，此制度為最低保證利率與投資報酬率連結。本文探討退休金給付標準為確定提撥制，當退休金的投資報酬率是根據其連結之股價指數的表現來計算時，股價指數報酬率的模型假設為馬可夫調控跳躍過程模型，考慮市場狀態與布朗運動項、跳躍項的跳躍頻率相關，即為Elliot et al. (2007) 的模型特例。使用1999年至2012年的道瓊工業指數與S&P 500指數的股價指數對數報酬率作為研究資料，採用EM演算法估計參數及SEM演算法估計參數共變異數矩陣。透過概似比檢定說明馬可夫調控跳躍過程模型比狀態轉換模型、跳躍風險下狀態轉換模型更適合描述股價指數報酬率變動情形，也驗證馬可夫調控跳躍過程模型具有描述報酬率不對稱、高狹峰及波動叢聚的特性。最後，假設最低保證利率為固定下，利用Esscher轉換法計算不同模型下型I保證之確定提撥制退休金的評價公式，從公式中可看出受雇人提領的退休金價值可分為政府補助與個人帳戶擁有之退休金兩部分。以執行敏感度分析探討估計參數對於馬可夫調控跳躍過程模型評價公式的影響，而型II保證之確定提撥制退休金的價值則以蒙地卡羅法模擬並探討其敏感度分析結果。 / Pension plan make people a guarantee life in their retirement. In order to ensure the appropriate amount of pension plan, government guarantees associated with pension plan which ties minimum rate of return guarantees and underlying asset rate of return. In this paper, we discussed the pension plan with defined contribution (DC). When the return of asset is based on the stock indices, the return model was set on the assumption that markov-modulated jump diffusion model (MMJDM) could the Brownian motion term and jump rate be both related to market states. This model is the specific case of Elliot et al. (2007) offering. The sample observations is Dow-Jones industrial average and S&P 500 index from 1999 to 2012 by logarithm return of the stock indices. We estimated the parameters by the Expectation-Maximization (EM) algorithm and calculated the covariance matrix of the estimates by supplemented EM (SEM) algorithm. Through the likelihood ratio test (LRT), the data fitted the MMJDM better than other models. The empirical evidence indicated that the MMJDM could describe the asset return for asymmetric, leptokurtic, volatility clustering particularly. Finally, we derived different model's valuation formula for DC pension plan with type-I guarantee by Esscher transformation under rate of return guarantees is constant. From the formula, the value of the pension plan could divide into two segment: government supplement and employees deposit made pension to their personal bank account. And then, we done sensitivity analysis through the MMJDM valuation formula. We used Monte Carlo simulations to evaluate the valuation of DC pension plan with type-II guarantee and discussed it from sensitivity analysis.

本文使用Esscher轉換法推導狀態轉換模型、跳躍獨立風險下狀狀態轉換模型及跳躍相關風險下狀態轉換模型的選擇權定價公式。藉由1999年至2011年道瓊工業指數真實市場資料使用EM演算法估計模型參數並使用概似比檢定得到跳躍相關風險下狀態轉換模型最適合描述報酬率資料。接著進行敏感度分析得知，高波動狀態的機率、報酬率的整體波動度及跳躍頻率三者與買權呈現正相關。最後由市場驗證可知，跳躍相關風險下狀態轉換模型在價平及價外的定價誤差皆是最小，在價平的定價誤差則略高於跳躍獨立風險下狀態轉換模型。 / In this paper, we derive regime switching model, regime switching model with independent jump and regime switching model with dependent jump by Esscher transformation. We use the data from 1999 to 2011 Dow-Jones industrial average index market price to estimate the parameter by EM algorithm. Then we use likelihood ratio test to obtain that regime switching model with dependent jump is the best model to depict return data. Moreover, we do sensitivity analysis and find the result that the probability of the higher volatility state , the overall volatility of rate of return , and the jump frequency are positively correlated with call option value. Finally, we enhance the empirical value of regime switching model with dependent jump by means of calculating the price error.