scholarly journals Detection of Heterogeneous Structures on the Gaussian Copula Model Using Projective Power Entropy

2013 ◽  
Vol 2013 ◽  
pp. 1-10 ◽  
Author(s):  
Akifumi Notsu ◽  
Yoshinori Kawasaki ◽  
Shinto Eguchi
Keyword(s):  
Gaussian Copula ◽  
Copula Model ◽  
Heterogeneous Structures

scholarly journals A fully pipelined probability density function engine for Gaussian Copula model

2014 ◽  
Vol 19 (2) ◽  
pp. 194-202 ◽  
Author(s):  
Huabin Ruan ◽  
Xiaomeng Huang ◽  
Haohuan Fu ◽  
Guangwen Yang
Keyword(s):  
Probability Density Function ◽  
Probability Density ◽  
Density Function ◽  
Gaussian Copula ◽  
Copula Model

An FPGA-Based Data Flow Engine for Gaussian Copula Model

2013 ◽  
Author(s):  
Huabin Ruan ◽  
Xiaomeng Huang ◽  
Haohuan Fu ◽  
Guangwen Yang ◽  
Wayne Luk ◽  
...  
Keyword(s):  
Data Flow ◽  
Gaussian Copula ◽  
Copula Model

scholarly journals A Diagnostic Test Based on a 9-Component Mixture Gaussian Copula Model

2021 ◽  
Vol 10 (4) ◽  
pp. 1
Author(s):  
A. Nanthakumar
Keyword(s):  
Infectious Disease ◽  
Markov Chain ◽  
Receiver Operating Characteristic Curve ◽  
Diagnostic Test ◽  
Operating Characteristic ◽  
Characteristic Curve ◽  
Gaussian Copula ◽  
Copula Model ◽  
Component Mixture ◽  
Operating Characteristic Curve

Here in this paper, we investigate the performance of a diagnostic test based on a mixture Gaussian Copula which incorporates a Markov Chain. Suppose that in the context of an infectious disease, there are three states; Susceptible (S), Infected (I), or Recovered (R). We compare the performance of this approach with the ROC (Receiver Operating Characteristic) Curve which is usually used in diagnostic studies.

scholarly journals Invariance Properties in the Dynamic Gaussian Copula Model

2017 ◽  
Vol 56 ◽  
pp. 22-41 ◽  
Author(s):  
Stéphane Crépey ◽  
Shiqi Song
Keyword(s):  
Gaussian Copula ◽  
Copula Model ◽  
Invariance Properties

Investigation for Synthetic Collateralized Debt Obligation

2006 ◽  
Vol 14 (1) ◽  
pp. 127-168
Author(s):  
Mi Ae Kim
Keyword(s):  
Default Probability ◽  
Gaussian Copula ◽  
Copula Model ◽  
Pricing Models ◽  
Hedging Strategy ◽  
Collateralized Debt Obligation ◽  
Collateralized Debt ◽  
Non Gaussian ◽  
Asset Correlation ◽  
Debt Obligation

Recently, domestic market participants have a growing interest in synthetic Collateralized Debt Obligation (CDO) as a security to reduce credit risk and create new profit. Therefore, the valuation method and hedging strategy for synthetic CDO become an important issue. However, there is no won-denominated credit default swap transactions, which are essential for activating synthetic CDO transaction‘ In addition, there is no transparent market information for the default probability, asset correlation, and recovery rate, which are critical variables determining the price of synthetic CDO. This study first investigates the method of estimating the default probability, asset correlation coefficient, and recovery rate. Next, using five synthetiC CDO pricing models‘ widely used OFGC (One-Factor Non-Gaussian Copula) model. OFNGC (One-Factor Non-Gaussian Copula) model such as OFDTC (One-Factor Double T-distribution Copula) model of Hull and White (2004) or NIGC (Normal Inverse Gaussian Copula) model of Kalemanova et al.(2005), SC<Stochastic Correlation) model of Burtschell et al.(2005), and FL (Forward Loss) model of Bennani (2005), I Investigate and compare three points: 1) appropriateness for portfolio loss distribution, 2) explanation for standardized tranche spread, 3) sensitivity for delta-neutral hedging strategy. To compare pricing models, parameter estimation for each model is preceded by using the term structure of iTraxx Europe index spread and the tranch spreads with different maturities and exercise prices Remarkable results of this study are as follows. First, the probability for loss interval determining mezzanine tranche spread is lower in all models except SC model than OFGC model. This result shows that all mαdels except SC model in some degree solve the implied correlation smile phenomenon, where the correlation coefficient of mezzanine tranche must be lower than other tranches when OFGC model is used. Second, in explaining standardized tranche spread, NIGC model is the best among various models with respect to relative error. When OFGC model is compared with OFDTC model, OFOTC model is better than OFGC model in explaining 5-year tranche spreads. But for 7-year or 10-year tranches, OFDTC model is better with respect to absolute error while OFGC model is better with respect to relative error. Third, the sensitivity sign of senior tranctle spread with respect to asset correlation is sometime negative in NIG model while it is positive in other models. This result implies that a long position may be taken by the issuers of synthet.ic COO as a correlation delta-neutral hedging strategy when OFGC model is used, while a short position may be taken when NIGC model is used.

Introducing an analytical solution and an improved one-factor gaussian copula model for the pricing of heterogeneous CDOs

2017 ◽  
Vol 04 (02n03) ◽  
pp. 1750038
Author(s):  
Xin Gao ◽  
Binlin Wu ◽  
Tobias Schäfer
Keyword(s):  
Analytical Solution ◽  
Analytical Model ◽  
Correlation Coefficients ◽  
Gaussian Copula ◽  
Copula Model ◽  
Collateralized Debt Obligations ◽  
Copula Models ◽  
Heterogeneous Model ◽  
Collateralized Debt ◽  
Individual Rate

This paper introduced an analytical solution and improved one-factor Gaussian copula models to the pricing of tranches of a Collateralized debt obligations (CDO) portfolio. Prices of CDO tranches are calculated and compared using the analytical model and different one-factor Gaussian copula models including a two-category heterogeneous model and a completely heterogeneous model that uses individual rate parameter and correlation coefficient for each reference entity in a CDO portfolio. When correlation among reference entities is low, the price calculated from the analytical model matches very well with the one-factor Gaussian copula models. However, as the correlation among reference entities increases, prices calculated using both the analytical solution and the homogeneous or two-category one-factor Gaussian copula models significantly deviate from the completely heterogeneous one-factor Gaussian copula model. This result verifies our belief that uniform parameters cannot completely capture all the heterogeneities in a CDO portfolio. Completely heterogeneous one-factor Gaussian copula model using individual rate parameters and correlation coefficients for each reference entities provides more reliable and accurate prices for structured securities.

INFORMATIONALLY DYNAMIZED GAUSSIAN COPULA

2013 ◽  
Vol 16 (02) ◽  
pp. 1350008 ◽  
Author(s):  
S. CRÉPEY ◽  
M. JEANBLANC ◽  
D. WU
Keyword(s):  
Credit Risk ◽  
State Space ◽  
Cox Model ◽  
Credit Derivatives ◽  
Point Of View ◽  
Gaussian Copula ◽  
Copula Model ◽  
Counterparty Risk ◽  
Gaussian Distributions ◽  
Augmented State

In order to dynamize the static Gaussian copula model of portfolio credit risk, we introduce a model filtration made of a reference Brownian filtration progressively enlarged by the default times. This yields a multidimensional density model of default times, where, as opposed to the classical situation of the Cox model, the reference filtration is not immersed into the enlarged filtration. In mathematical terms this lack of immersion means that martingales in the reference filtration are not martingales in the enlarged filtration. From the point of view of financial interpretation this means default contagion, a good feature in the perspective of modeling counterparty wrong-way risk on credit derivatives. Computational tractability is ensured by invariance of multivariate Gaussian distributions through conditioning by some components, the ones corresponding to past defaults. Moreover the model is Markov in an augmented state-space including past default times. After a discussion of different notions of deltas, the model is applied to the valuation of counterparty risk on credit derivatives.

Sign in / Sign up

Export Citation Format

Share Document