none

資產證券化源自1970年代，第一筆擔保債權憑證交易自1988年出現在美國，然後在歐美迅速發展，目前已成為重要的之債券市場。台灣金融產業發展正值轉型期，銀行除面對低利率帶來經營壓力之外，同時亦需規避評等較差之企業貸款的信用風險，而保險業者在低利率時代來臨卻無良好報酬之投資標的可供投資。因此，此環境乃為推動證券化之良好契機。自1997年發生東南亞金融危機，乃至1998年韓國的亞洲金融危機，造成許多跨國企業紛紛裁員、關廠、甚至倒閉，造成一連串的金融危機連鎖效應。因此，公司間或產業間之榮枯是相互關聯的，且均會受總體經濟因素所影響。是以，近年來除信用風險亦成為近年來財務領域上重要議題。理論或實證上，當多個標的資產之信用衍生性商品被加以開發，並用來管理信用風險的時候，需考慮多個標的資產間的違約相關性，方能準確地衡量信用風險。故在信用風險管理與信用衍生性商品的評價中，違約相關性的估計與考量顯得格外重要。結構式或縮減式模型在發展違約相關性的多變數模型中是困難的，因為其衍生性商品價值的理論推導繁複或其數值計算是相當費時。本文在多標的資產之信用風險評價模型中，透過適當個別資產之邊際違約機率與Copula函數之選擇，及其相關參數之估算，即可快速求算具違約相關性之多變數聯合機率函數，以利擔保債權憑證(CDO)之評價。因此，本文針對BET、Copula、Factor Copulas等三種方法與分析架構做一剖析，再以國內153家上市公司所發行無擔保債券作為連結標的之擔保債權憑證為例，進行模擬實證並分析結果。

Two important dimensions of work related choices are work location and working hours. Telecommuting (working from home or any convenient place instead of commuting to the conventional working place) can potentially have a substantial impact on traffic demand distribution on a particular day by means of its replacement and displacement effect. Consequently, it is of interest to analyze the effect and extent of telecommuting adoption across the labor force. This study proposes a copula based joint discrete multinomial-duration model of choice accommodating the two dimensions of work related choices: work arrangement and aggregated duration of telecommuting episodes on a particular day. In the econometric model telecommuting episodes are defined so as the duration is at least 30 minutes and only home-based telecommuting is considered and sample is drawn from the ATUS, 2007 data. The results from the estimated model show that gender, higher-education, responsibility for child-care, family ties act as driving forces for adopting telecommuting. The sign of the Gaussian copula parameter or dependency parameter implies that the unobserved factors act in opposite direction on the two dimensions: work arrangement choice and aggregated telecommuting episode duration. / text

Credit risk and market risk have already been explored intensively and the reliable models of credit risk and market risk have also been developed progressively. As to financial institution, how to control credit risks and venture capital to count and withdraw the implementation with new Basel capital protocol, will concern the competitiveness of the financial institution. This study try to find a method pricing the CDO (Collateralized Debt Obligation) based on Macroeconomic and financial ratio credit model. For the various approaches to CDO valuation, the most widely accepted is the Copula approach. The Copula approach is considered suitable for describing default correlation. Combining with Monte Carlo Simulation, it can price CDO effectively.

This dissertation includes two essays. In the first essay, I proposed an alternative estimator for multivariate densities. This estimator can be characterized as a transformation based estimator. The first stage estimates each marginal density separately. In the second stage, the joint density of estimated marginal cumulative distribution functions (CDF) are approximated by the exponential series estimator. The final estimate is then obtained as the product of the marginal densities and the joint density estimated in the second stage. Extensive Monte Carlo studies show the proposed estimator outperforms kernel estimators in joint density and tail distribution estimation. An illustrative example on estimating the conditional copula density between S & P 500 and FTSE 100 given Hangseng and Nikkei 225 is also discussed. In the second essay, I extended the semiparametric model by Chen and Fan [X. Chen, Y. Fan, Estimation of copula-based semiparametric time series models, Journal of Econometrics 130 (2006) 307-335], and studied a class of univariate copula-based nonparametric stationary Markov models in which the copulas and the marginal distributions are estimated nonparametrically. In particular, I focused on the stationary Markov process of order 1 with continuous state space because it has the beta-mixing property for the analysis of weakly dependent processes. The copula density functions for time series models are approximated by the series estimate on sieve spaces. In this study, a finite dimensional linear space spanned by a sequence of power functions is treated as the sieve space where the estimation space of the copula density function is based. This sieve series estimator can be characterized as the exponential series estimator under mild smoothness conditions. By using the beta-mixing properties, I showed that the copula density function approximated by the exponential series estimator for stationary first-order Markov processes has the same convergence rate as the i.i.d. data. The Monte Carlo simulations show that the proposed estimator outperforms the kernel estimator in the conditional density estimation, except for the Frank copula-based Markov model. In addition, the proposed estimator considerably dominates the the kernel estimator when used in the one-step-ahead forecast.

Ein Ausfallkorb (Default Basket, Basket Default Swap, BDS) ist die Bündelung einer relativ geringen Anzahl einzelner Kreditpositionen. Der Sicherungsgeber (Investor) verpflichtet sich, den i-ten Forderungsausfall zu übernehmen und wird als ith-to-default-Käufer bezeichnet. Da es sich um die Bündelung einer relativ geringen Anzahl von, möglicherweise sehr heterogenen Kreditpositionen handelt, lassen sich herkömmliche Modellierungsansätze aus dem Kreditrisiko nicht direkt zur Risikobewertung anwenden. Der vorliegende Beitrag stellt deshalb Möglichkeiten vor, eine Risikobewertung für Ausfallkörbe vorzunehmen. Der Modellierungsansatz über das Ein-Faktormodell ist dabei sehr stark an die typische Kreditrisikomodellierung angelehnt, weicht aber in einigen Punkten auch erheblich davon ab. Zentrales Anliegen dieses Artikels ist es daher, die wesentlichen Mechanismen zu verdeutlichen, welche die Risikobewertung in diesem Modell beeinflussen. Hierbei wird insbesondere das Zusammenspiel von Abhängigkeitsstruktur (in Form der Korrelation), Ausfallwahrscheinlichkeiten der Einzelpositionen und den Ausfallwahrscheinlichkeiten für die Risikogeber betrachtet.

A measure of concordance, $\kappa$, is of polynomial type if and only if $\kappa (tA+(1-t)B)$ is a polynomial in $t$ where $A$ and $B$ are 2-copulas. The degree of such a type of measure of concordance is simply the highest degree of the polynomial associated with $\kappa$. In previous work [2], [3], properties of measures of concordance preserving convex sums (equivalently measures of concordance of polynomial type degree one) were established; however, a characterization was not made. Here a characterization is made using approximations involving doubly stochastic matrices. Other representations are provided from this characterization leading naturally to two interpretations of degree one measures of concordance. The existence of a family of measures of concordance of polynomial type having higher degree generated by a certain family of Borel measures on $(0,1)^{2n}$ is also shown. The representation of this family immediately leads to a probabilistic interpretation for all finite measures in $d_n$. Also, higher degree analogs of commonly known degree one measures of concordance are given as examples. For the degree 2 case in particular, we see there is no finite measure in $d_2$ generating Kendall's tau. Finally, another family of measures of concordance is given containing those generated by finite measures in $d_2$ as well as Kendall's tau.

The primary aim of this thesis is the elucidation of covariate effects on the dependence structure of random variables in bivariate or multivariate models. We develop a unified approach via a conditional copula model in which the copula is parametric and its parameter varies as the covariate. We propose a nonparametric procedure based on local likelihood to estimate the functional relationship between the copula parameter and the covariate, derive the asymptotic properties of the proposed estimator and outline the construction of pointwise confidence intervals. We also contribute a novel conditional copula selection method based on cross-validated prediction errors and a generalized likelihood ratio-type test to determine if the copula parameter varies significantly. We derive the asymptotic null distribution of the formal test. Using subsets of the Matched Multiple Birth and Framingham Heart Study datasets, we demonstrate the performance of these procedures via analyses of gestational age-specific twin birth weights and the impact of change in body mass index on the dependence between two consequent pulse pressures taken from the same subject.

Au cours de ces vingt dernières années le secteur de la microfinance au niveau global a crû de manière exponentielle. Il a permis à des millions de personnes à faibles revenus et exclues du secteur de la finance traditionnelle, de mieux lisser leur consommation et d'investir pour améliorer leur bien-être. Cependant au cours de cette même période, le secteur de la microfinance a aussi été frappé par des crises de repaiement et un certain nombre d'institutions de microfinance (IMFs) ont vu leur taux de défaut augmenter de manière alarmante. Il est ainsi crucial d'avoir une connaissance approfondie des determinants du risque au niveau macro et micro. Cette thèse, apporte une contribution dans ce sens en étudiant les risques de crédit au niveau du secteur et au niveau des empreunteurs en microfinance. Dans le premier chapitre, nous suggérons une nouvelle mesure de risque systémique pour le marché du microcrédit. Nous explorons ensuite les déterminants de ce risque pour 37 pays de 2000 à 2014. Pour ce faire, nous modélisons, la distribution jointe de la qualité du portefeuille de toutes les IMFs dans un pays donné en utilisant une copule équidépendante semi-paramétrique. Notre mesure est basée sur l'idée qu'une plus forte dépendance de la qualité du portefeuille dans un pays est un indicateur de fagilité dans le secteur. Nous montrons qu'après avoir controlé pour les effets pays, la concurrence, le niveau de pénétration du marché, la commercialisation, la croissance excessive du secteur ainsi qu'un fort taux d'intérêt augmente la fragilité du secteur. Dans le second chapitre, nous utilisons des données de la plateforme de microfinance Kiva.org pour évaluer le taux de défaut des prêts que les IMFs postent sur la platforme. Nous identifions, d'abord que les prêts de plus longue durée ont un taux de défaut plus élevés, deuxièmement, que le taux de défaut baisse avec l'âge de l'emprunteur et finalement que les délais de grâce augmentent le taux de défaut pour les prêts de groupe avec une fréquence de remboursement mensuelle.L'évaluation des politiques publiques est particulière difficile notamment lorsque celles-ci affectent un grand nombre de personnes et ont des effets non-marginaux. Le troisième chapitre de cette thèse apporte une contribution à cette littérature en estimant l'effet d'une politique éducative visant à donner plus de choix d'écoles aux ménages sur le prix de la qualité des lycées en Corée du Sud. Plus précisément, nous utilisons une approche hédonique pour estimer l'effet de la politique éducative sur le prix d'équilibre de la qualité des lycée à Séoul. Notre stratégie d'identification repose sur des effets spatiaux précis comme des gratte-ciel d'appartements à Séoul. C'est effets nous permettent d'estimer le prix d'équilibre de la qualité des lycées avant et après la politique sans avoir à se reposer sur la politique elle-même comme stratégie d'identification. Ainsi, nous mitigeons le biais dû au fait que les ménages se répartissent en fonction de leur préférence pour les biens localement distribués. Nous estimons que cette politique de choix d'école diminue le prix d'équilibre de la qualité des lycées de USD 26 871 en moyenne. / The last 20 years have witnessed an exponential growth of the microfinance sector at the global level. In so doing, it has provided millions of poor people previously excluded from the traditional financial sector a better way to smooth their consumption and to invest to improve their livelihood.However, over the same period, the microfinance sector has also been hit by a series of repayment crises, and many microfinance institutions have seen an increase in their default rates at the microlevel. It is therefore important to have a better understanding of the determinants of risk in the microfinance sector. In this thesis, we address this issue by exploring the determinant of risk at the macro and microlevel. In the first chapter we introduce a new copula-based measure of systemic credit risk in microcredit markets and explore its determinants for 37 countries from 2000 to 2014. We model the joint distribution of portfolio quality of all microfinance institutions (MFIs) in a givencountry using an equidependent copula and an empirical density for the marginal. Our methodology is based on the idea that a higher level of dependence among MFIs in a given country is an indicator of potential fragility of the sector.We show that, after controlling for country effects, measures of competition, market penetration, commercialization and excessive growth of the sector, as well as the level of interest rates charged by MFIs in a country all increase the fragility of the microfinance sector, while lending to more women tends to reduce systemic risk. With this measure of systemic risk, we further compute the probability that a proportion of at least 20% of MFIs in a country are in serious financial distress and our simulations capture a progressive increase in risk in many countries that experienced a repayment crisis. In the second chapter, we use data from the microfinance crowdfunding platform Kiva to analyze the determinants of microfinance borrower defaults in a broad range of microfinance institutions in different countries. We use loan and borrower characteristics, downloaded from Kiva.org, to evaluate the repayment rates of the loans, that MFIs post on the platform. We identify, firstly that loans with longer durations have higher default, secondly that default rates decrease with the age of the borrower. Finally we find that grace periods increase default rates for group loans with monthly repayment frequency while the effect goes the other way with a weekly repayment frequency.The evaluation of public policies is challenging, particularly when these policies are meant to have wide scale and non-marginal effects. The third chapter of this thesis contribute to this literature by estimating the effect of the implementation of an open school choice program in South Korea. To be more specific we use a hedonic approach to estimate the effect of a high school choice program on the equilibrium price of high school quality in Seoul. Our identification strategy makes use of fine scale spatial effects such as clusters of skyscrapers apartment buildings and allows us to estimate the equilibrium price of high school quality before and after the school choice program without relying on the policy itself as an instrument.By allowing the spatial effects to differ before and after the policy, this strategy also mitigates the bias due to the sorting behavior of households that is likely to happen in the case of policies of such scales. We find that the policy decreases the equilibrium price of high school quality by USD 26,871 on average.

The primary aim of this thesis is the elucidation of covariate effects on the dependence structure of random variables in bivariate or multivariate models. We develop a unified approach via a conditional copula model in which the copula is parametric and its parameter varies as the covariate. We propose a nonparametric procedure based on local likelihood to estimate the functional relationship between the copula parameter and the covariate, derive the asymptotic properties of the proposed estimator and outline the construction of pointwise confidence intervals. We also contribute a novel conditional copula selection method based on cross-validated prediction errors and a generalized likelihood ratio-type test to determine if the copula parameter varies significantly. We derive the asymptotic null distribution of the formal test. Using subsets of the Matched Multiple Birth and Framingham Heart Study datasets, we demonstrate the performance of these procedures via analyses of gestational age-specific twin birth weights and the impact of change in body mass index on the dependence between two consequent pulse pressures taken from the same subject.