A Bayesian simulation approach to inference on a multi-state latent factor intensity model. (English) Zbl 1274.62406
Summary: The influence of economic conditions on the movement of a variable between states (for example a change in credit rating from A to B) can be modelled using a multi-state latent factor intensity framework. Estimation of this type of model is, however, not straightforward, as transition probabilities are involved and the model contains a few highly analytically intractable distributions. In this paper, a Bayesian approach is adopted to manage the distributions. The innovation in the sampling algorithm used to obtain the posterior distributions of the model parameters includes a particle filter step and a Metropolis-Hastings step within a Gibbs sampler. The feasibility and accuracy of the proposed sampling algorithm is supported with a few simulated examples. The paper contains an application concerning what caused 1049 firms to change their credit ratings over a span of ten years.
MSC:
| 62H25 | Factor analysis and principal components; correspondence analysis |
| 62F15 | Bayesian inference |
| 62P05 | Applications of statistics to actuarial sciences and financial mathematics |
| 91G50 | Corporate finance (dividends, real options, etc.) |
Keywords:
auxiliary particle filter; latent factor model; non-linear non-Gaussian state space model; transitions in credit ratingsReferences:
| [1] | Aguilar, Bayesian dynamic factor models and portfolio allocation, J. Bus. Econom. Statist. 18 pp 338– (2000) · doi:10.2307/1392266 |
| [2] | Bangia, Ratings migration and the business cycle with applications to credit portfolio stress testing, J. Banking and Finance 26 pp 445– (2002) · doi:10.1016/S0378-4266(01)00229-1 |
| [3] | Bauwens , L. Hautsch , N. 2003 Dynamic latent factor models for intensity processes Catholic University of Louvain |
| [4] | Bauwens, The stochastic conditional duration model: a latent variable model for the analysis of financial durations, J. Econometrics 119 pp 381– (2004) · Zbl 1282.91236 · doi:10.1016/S0304-4076(03)00201-X |
| [5] | Cai, Metropolis-Hastings algorithms with adaptive proposals, Statist. Comput. 18 pp 421– (2008) · doi:10.1007/s11222-008-9051-5 |
| [6] | Chib, Marginal likelihood from the Gibbs output, J. Amer. Statist. Assoc. 90 pp 1313– (1995) · Zbl 0868.62027 · doi:10.2307/2291521 |
| [7] | Chib, Marginal likelihood from the Metropolis-Hastings output, J. Amer. Statist. Assoc. 96 pp 270– (2001) · Zbl 1015.62020 · doi:10.1198/016214501750332848 |
| [8] | Chib, Markov chain Monte Carlo methods for stochastic volatility models, J. Econometrics 108 pp 281– (2002) · Zbl 1099.62539 · doi:10.1016/S0304-4076(01)00137-3 |
| [9] | Doucet, On sequential Monte Carlo sampling methods for Bayesian filtering, Statist. Comput. 10 pp 197– (2000) · doi:10.1023/A:1008935410038 |
| [10] | Duffie, Multi-period corporate default prediction with stochastic covariates, J. Financial Economics 83 pp 635– (2007) · doi:10.1016/j.jfineco.2005.10.011 |
| [11] | Fiorentini, Likelihood-based estimation of latent generalised ARCH Structures, Econometrica 72 pp 1481– (2004) · Zbl 1091.62071 · doi:10.1111/j.1468-0262.2004.00541.x |
| [12] | Gelfand, Bayesian model choice: asymptotics and exact calculations, J. Roy. Statist. Soc. Ser. B 56 pp 501– (1994) · Zbl 0800.62170 |
| [13] | Geweke, Bayesian Statistics pp 169– (1992) |
| [14] | Geweke, Using simulation methods for Bayesian econometric models: inference, development, and communciation, Econometric Rev. 18 pp 1– (1999) · Zbl 0930.62105 · doi:10.1080/07474939908800428 |
| [15] | Hawkes, Spectra of some self-exciting and mutually exciting point processes, Biometrika 58 pp 83– (1971) · Zbl 0219.60029 · doi:10.1093/biomet/58.1.83 |
| [16] | Hu, The estimation of transition matrices for sovereign credit ratings, J. Banking and Finance 26 pp 1383– (2002) · doi:10.1016/S0378-4266(02)00268-6 |
| [17] | Kass, Bayes factors, J. Amer. Statist. Assoc. 90 pp 773– (1995) · Zbl 0846.62028 · doi:10.2307/2291091 |
| [18] | Koopman, Business and default cycles for credit risk, J. Appl. Econometrics 20 pp 311– (2005) · doi:10.1002/jae.833 |
| [19] | Koopman, The multi-state latent factor intensity model for credit rating transitions, J. Econometrics 142 pp 399– (2008) · Zbl 1418.62383 · doi:10.1016/j.jeconom.2007.07.001 |
| [20] | McNeil, Bayesian inference for generalized linear mixed models of portfolio credit risk, J. Empir. Finance 14 pp 131– (2007) · doi:10.1016/j.jempfin.2006.05.002 |
| [21] | Newton, Approximate Bayesian inference by the weighted likelihood bootstrap (with Discussion), J. Roy. Statist. Soc. Ser. B. 56 pp 3– (1994) · Zbl 0788.62026 |
| [22] | Nickell, Stability of rating transitions, J. Banking and Finance 10 pp 423– (2000) |
| [23] | Pitt , M. 2002 Smooth particle filters for likelihood maximisation |
| [24] | Pitt, Filtering via simulation: auxiliary particle filter, J. Amer. Statist. Assoc. 94 pp 590– (1999) · Zbl 1072.62639 · doi:10.2307/2670179 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
