Multiperiod Corporate Default Prediction with the Partially-Conditioned Forward Intensity

Duan, Jin‐Chuan; Fülöp, András

doi:10.2139/ssrn.2151174

Cited by 17 publications

(26 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Certain aspects of the model are taken from Duan and Fulop (2013). This model allows PD forecasts to be made at a range of horizons.…”

Section: Model Descriptionmentioning

confidence: 99%

“…The parameter estimation for the NS functions is based on a new numerical method (a pseudo-Bayesian SMC technique) developed in a working paper by Duan and Fulop (2013). The remainder of this section details the new parameter estimation.…”

Section: I3 Parameter Estimationmentioning

confidence: 99%

“…show that this approximation by NS functions does not negatively affect prediction performance. The RMI implementation follows Duan and Fulop (2013) to impose the functional restriction during the estimation as opposed to the method used in of fitting the curve after parameter estimates have been obtained. This is done for two reasons.…”

Section: I31 Smoothed Parametersmentioning

confidence: 99%

“…Moreover, the number of parameters can be greater than 40 if there are more than 12 covariates. The RMI implementation follows Duan and Fulop (2013) who use the SMC pseudo-Bayesian method for estimation and self-normalized statistics for inference.…”

Section: I32 Parameter Estimation By Smcmentioning

confidence: 99%

“…The following exposition closely follows the Appendix in Duan and Fulop (2013). It is important to note that the RMI implementation uses the model described in Duan and Fulop (2013), which does not contain any latent frailty or partial conditioning variable, and hence is technically much simpler in parameter estimation.…”

Section: I32 Parameter Estimation By Smcmentioning

confidence: 99%

See 4 more Smart Citations

NUS-RMI Credit Research Initiative Technical Report – Version: 2014 Update 1

staff¹

2014

Global Credit Review

View full text Add to dashboard Cite

“…Certain aspects of the model are taken from Duan and Fulop (2013). This model allows PD forecasts to be made at a range of horizons.…”

Section: Model Descriptionmentioning

confidence: 99%

Section: I3 Parameter Estimationmentioning

confidence: 99%

Section: I31 Smoothed Parametersmentioning

confidence: 99%

Section: I32 Parameter Estimation By Smcmentioning

confidence: 99%

Section: I32 Parameter Estimation By Smcmentioning

confidence: 99%

See 3 more Smart Citations

NUS-RMI Credit Research Initiative Technical Report – Version: 2014 Update 1

staff¹

2014

Global Credit Review

View full text Add to dashboard Cite

Sequential Monte Carlo optimization and statistical inference

Duan

2022

WIREs Computational Stats

View full text Add to dashboard Cite

Sequential Monte Carlo (SMC) is a powerful technique originally developed for particle filtering and Bayesian inference. As a generic optimizer for statistical and nonstatistical objectives, its role is far less known. Density-tempered SMC is a highly efficient sampling technique ideally suited for challenging global optimization problems and is implementable with a somewhat arbitrary initialization sampler instead of relying on a prior distribution. SMC optimization is anchored at the fact that all optimization tasks (continuous, discontinuous, combinatorial, or noisy objective function) can be turned into sampling under a density or probability function short of a norming constant. The point with the highest functional value is the SMC estimate for the maximum. Through examples, we systematically present various density-tempered SMC algorithms and their superior performance vs. other techniques like Markov Chain Monte Carlo. Data cloning and k-fold duplication are two easily implementable accuracy accelerators, and their complementarity is discussed. The Extreme Value Theorem on the maximum order statistic can also help assess the quality of the SMC optimum. Our coverage includes the algorithmic essence of the density-tempered SMC with various enhancements and solutions for (1) a bi-modal nonstatistical function without and with constraints, (2) a multidimensional step function, (3) offline and online optimizations, (4) combinatorial variable selection, and (5) noninvertibility of the Hessian. This article is categorized under: Statistical and Graphical Methods of Data Analysis > Monte Carlo Methods Algorithms and Computational Methods > Stochastic Optimization Algorithms and Computational Methods > Integer Programming K E Y W O R D S combinatorial optimization, data cloning, density tempering, extreme value 1 | INTRODUCTION Sequential Monte Carlo (SMC) is, as the name suggests, a family of Monte Carlo sampling techniques that sequentially adapt to the target density/probability function by moving/concentrating the simulated sample to the most desirable

show abstract

Dynamic Portfolio Credit Risk and Large Deviations

Juneja

2017

Econophysics and Sociophysics: Recent Progress and Future Directions

View full text Add to dashboard Cite

Multiperiod Corporate Default Prediction with the Partially-Conditioned Forward Intensity

Cited by 17 publications

References 18 publications

NUS-RMI Credit Research Initiative Technical Report – Version: 2014 Update 1

NUS-RMI Credit Research Initiative Technical Report – Version: 2014 Update 1

Sequential Monte Carlo optimization and statistical inference

Dynamic Portfolio Credit Risk and Large Deviations

Contact Info

Product

Resources

About