Numerical Modeling of Dependent Credit Rating Transitions with Asynchronously Moving Industries
 1.2k Downloads
 1 Citations
Abstract
Two models of dependent credit rating migrations governed by industryspecific Markovian matrices, are considered. Caused by macroeconomic factors, positive and negative unobserved tendencies, encoded as values “1” or “0” of the corresponding variables, modify the transition probabilities and render the evolutions dependent. They are neither synchronized across industry sectors, nor over credit classes: an upswing in some of them can coexist with a decline of the rest. The models are tested on Standard and Poor’s data. MATLAB optimization software and maximum likelihood estimators are used. Obtained distributions of the hidden variables demonstrate that the considered industries migrate asynchronously trough credit classes. Since downgrading probabilities are less affected by the unobserved tendencies, estimated by MonteCarlo simulations distributions of defaults, exhibit lighter, than for the known coupling models, tails for schemes with asynchronously moving industries. Moreover, the lightest tails were obtained in the case of industryspecific transition matrices.
Keywords
Macroeconomic factor Markov process Loss distribution Maximum likelihood Credit rating MonteCarlo simulations CorrelationJEL Classification
C44 C61 G171 Introduction
Within the CreditMetrics approach, the study of changes in the credit quality of debtors through time is a corner stone, see Gupton et al. (1997). While the credit rating of each of them evolves as a time homogeneous Markov chain, in order to model the joint distribution of a pool of debtors, a coupling scheme can be suggested. Then, introducing dependence among the migrations, the evolution of every debtor in the pool can be represented as a randomization of an idiosyncratic move and a common component. In particular, the model by Kaniovski and Pflug (2007) assumes a single common component for all debtors belonging to a credit class, while in the modification by Wozabal and Hochreiter (2012) common components are debtorspecific. Boreiko et al. (2015) study an intermediate situation when a common component remains the same for all debtors characterized by a combination of a credit class and an industry sector. In all three cases, the distribution of a common component depends on an unobserved binary tendency variable. This variable indicates whether the overall state of the economy is favorable or not for debtors belonging to the credit class in question.
In the coupling schemes quoted above, positive or negative tendencies affect uniformly all debtors having the same credit rating, regardless of their industry sectors. In other words, it is assumed that the microeconomic factors influence all industry sectors in the same way. Let us label this pattern of tendency variables as synchronous evolution of industries.
In this paper, asynchronously moving industries are analyzed. That is, tendency variables are neither synchronized across industry sectors, nor over credit classes: favorable conditions for some of them can coexist with adversities for the rest. This assumption could account better for the observed variability of the strength and of the direction of the macroeconomic factors across industry sectors. In addition, this setting allows to implement industryspecific transition matrices. This departure from the standard CreditMetrics approach may allow to model better the distributions governing credit rating migrations. For example, while estimating default correlations, default frequencies that vary across industry sectors would be a clear indication that the same Markovian matrix cannot govern migrations of all debtors. The corresponding multidimensional Markov credit rating processes and parameters defining them are specified in Sects. 2 and 3. The estimators are described in Sect. 4.
Evaluation of losses generated by a portfolio is a fundamental task of credit risk analysis. Typically the losses are quantified in terms of defaults. A range of approaches has been proposed in this case. See among others Li (2000), Nagpal and Bahar (2001), Bangia et al. (2002), Hull and White (2004), McNeil and Wendin (2007), Frydman and Schuermann (2008), Korolkiewicz and Elliott (2008), Stefanescu et al. (2009), Xing et al. (2012), ChorośTomczyk et al. (2013). In Sect. 7, VaR and CVaR of the distribution of the number of defaults are estimated and compared for several parameterizations of the two coupling schemes introduced here.
In order to test numerically the suggested models, a Standard and Poor’s (S&P’s) data set covering OECD (Organization for Economic Cooperation and Development) countries is used. A pool of debtors mimicking the portfolio generating the Dow Jones iTraxx EUR index is considered. The input data are characterized in Sect. 5.
Maximum likelihood estimates were obtained by two MATLAB constrained optimization programs, the interior point algorithm (IP) and the sequential quadratic programming method (SQP). The estimates demonstrate that the considered industries evolve, in fact, asynchronously. See Sect. 6. Some implications of the asynchronicity phenomenon for credit risk assessment are presented in Sect. 7. There are Monte–Carlo estimates for VaR and CVaR of the loss distribution and bounds for caused by macroeconomics factors variation of credit rating transition probabilities. Section 8 summarizes the obtained results. Appendix 1 contains estimated parameters. Given in Appendix 2 formulas for partial derivatives allow to accelerate substantially computations.
2 Credit Rating Migration Process

the evolutions of debtors through credit classes and industry sectors are dependent;

the corresponding random process of credit rating transitions is time homogeneous and every individual migration in industry sector s is governed by the same Markovian transition matrix \(P^{(s)}\).
The evolution of the whole portfolio is captured by a random process \(\vec {X}(t)=(X_{1}(t),X_{2}(t),\ldots ,X_{\mathcal {N} (1)}(t))\) whose components are stochastically dependent. Denote by s(n) the industry sector of debtor n. The rating randomly changes in time, becoming \(X_{n}(2)\) at time \(t=2\), while the assignment to sector s(n) remains the same. Since the distributions in question are timehomogeneous, it is enough to consider a transition from time \(t=1\) to time \(t=2\).
Equation (1) is similar to what is used in common factor models. See, for example, Hull and White (2004). However, unlike them, the coefficients \(\delta _n\) are stochastic in (1). As a consequence, the distribution of \(X_{n}(2)\) is a mixture of distributions of \(\xi _{n}\) and \(\eta _{n}\) with weights \(q_{X_{n}(1),s(n)}\) and \(1q_{X_{n}(1),s(n)}\), rather than being the convolution of them, as in the case of the common factor approach where the corresponding weights are deterministic.
Let us describe a dependence structure among common components. Denote by \(\{0,1\}^{MS}\) the set of all vectors with MS coordinates, each 0 or 1. Let \(\pi (\cdot )=\{\pi (\vec {\chi }),\quad \vec {\chi }\in \{0,1\}^{MS}\}\) be a probability distribution. A tendency vector \(\vec {\Pi }=(\Pi _1,\dots ,\Pi _{MS})\) is a random vector whose distribution is \(\pi (\cdot )\). Coordinates \(\Pi _i\) are not observable. They are termed as tendency, hidden or latent variables.
Let \(\vec {\chi }=(\chi _1,\ldots ,\chi _{MS})\) be a realization of a tendency vector. Its coordinate \(\chi _{M(s1)\,+\,i}\), a realization of the tendency variable \(\Pi _{M(s1)\,+\,i}\), affects the evolution of debtors from credit class i and industry sector s. The corresponding mechanism is described next.
When \(\chi _{M(s1)\,+\,i}=1\), all of the random variables \(\eta _n\), such that \(X_n(1)=i\) and \(s(n)=s\), cannot assume values larger than i. If credit class migrations of every debtor belonging to credit class i and industry sector s had been governed exclusively by the corresponding \(\eta _n\), this would have meant that the credit rating of such debtors cannot worsen. For this reason, the situation when \(\chi _{M(s1)\,+\,i}=1\) is termed as a nondeteriorating tendency for them. If \(\chi _{M(s1)\,+\,i}=0\), all of the random variables \(\eta _n\), such that \(X_n(1)=i\) and \(s(n)=s\), take on exclusively values exceeding i. Had it been determined exclusively by the common component, the credit rating would have worsen in this case. Hence, this is a deteriorating tendency.
There are two specifications for \(\{\eta _n\}\) termed as coupling schemes one and two in what follows next. In the first case, common components are debtorspecific. That is, conditional on \(\vec {\Pi }\), \(\{\eta _n\}\) are independent in n. In the second case, the same common component applies to all debtors characterized by a combination of an industry sector and a credit class. More precisely, given \(\vec {\Pi }\), random variables \(\eta _n\) and \(\eta _l\) are stochastically independent if \(s(n)\not =s(l)\) or \(X_n(1)\not =X_l(1)\), while \(\eta _n= \eta _l\) if \(s(n)=s(l)\) and \(X_n(1)=X_l(1)\).
3 Parameters of the Model

\(M\times (M+1)\) Markovian matrices \(P^{(s)}\), \(s=1,2,\ldots ,S\);

a probability distribution \(\pi (\cdot )\) on \(\{0,1\}^{MS}\);

an \(M\times S\) matrix Q formed by probabilities of success \(q_{i,s}\) of Bernoulli random variables in (1).
Since rating agencies report their Markovian transition matrices, conventionally a matrix P, common for all debtors, was assumed to be known and all estimation efforts concentrated on finding Q and \(\pi (\cdot )\). Here, following this tradition, matrices \(P^{(s)}\) are supposed to be given, while Q and \(\pi (\cdot )\) have to be estimated.
Wozabal and Hochreiter (2012) suggested maximum likelihood estimates for these parameters and a heuristic global search method for finding them. Kaniovski and Pflug (2007), for a given P, identified a distribution \(\pi (\cdot )\), satisfying relations (3), by solving a quadratic programming problem. Conceptually, they searched for a string distribution that is “closest” in the least squares sense to the one with independent tendency variables.
4 Likelihood Functions and Optimization Problems
The unknown parameters were estimated by maximizing logarithms of the likelihood functions subject to linear constraints, identical for both coupling schemes.
If the Markovian matrices are identical for all industries and the additional constraint (6) holds true, then the first coupling scheme is equivalent to the setting of Wozabal and Hochreiter (2012), while the second coupling scheme corresponds to the setting introduced in Boreiko et al. (2015). The likelihood functions given here can be obtained by a minor modification of the respective arguments in these papers.
If \(p_{i,j}^{(s)}=0\) for some i and j, then \(I^t(s,m_1,m_2)=0\) as well. Consequently, such terms are excluded from consideration. Also, containing no unknowns, the term I cannot affect a solution of the above maximization problems.
Estimating parameters of a coupled Markov chain model from real data, one deals with a mixture of multinomial distributions. This class of statistical problems is known to imply multiple solutions. A detailed analysis of these problems is given in Allman et al. (2009). However, in practical applications no complications seem to arise. See CarreiraPerpiñán and Renals (2000).
5 Input Data
A S&P’s data set covering companies from 30 OECD countries for the period from 1991 through 2013 was considered.
An illustrative example has to deal with a portfolio that is known in everyday practice of risk management. However, the number of industry sectors involved cannot be large for two reasons. First, there must be enough migrations in each industry sector. For example, for estimating the respective Markovian matrix. Second, the dimension of the optimization problem and, consequently, its run time of an optimization algorithm should not explode. Note that Q contains \(M\times S\) entries, while a distribution \(\pi (\cdot )\) is nested on \(2^{M\times S}\) sample points. Hence, the total number of unknowns is \(M\times S+2^{M\times S}\).
For example, in September of 2013, in the portfolio generating the Dow Jones iTraxx EUR market index investment grade debtors belonging to five industry sectors were involved: 1—auto and industrial; 2—consumer; 3—energy with utilities; 4—finance and insurance; 5—telecommunications, media and technology. That is, \(S=5\). Since there are two nondefault credit classes, investment grade and noninvestment grade debtors, \(M=2\). Then the total number of unknowns is \(10+2^{10}=1034\). Investment grade debtors are characterized by S&P’s ratings from AAA to BBB, while noninvestment grade ones occupy the ratings from BB and downward. Indexes 1, 2 and 3 refer to an investment grade, a noninvestment grade and a defaulted debtor, respectively.
With this choice of parameters, using finitedifference approximations of partial derivatives, the corresponding MATLAB software required typically 4–8 hours in order to find a solution. Evaluating the respective derivatives analytically, the runtime reduces drastically. In particular, to a couple of minutes in the case of IP method with first derivatives given in Appendix 2.
Keeping in mind the possibility of multiple solutions, a variety of initial approximations have been tried, including the use of a solution obtained by one of the methods as a starting point for the other one. In all cases the results reported here were identical for both MATLAB algorithms, IP and SQP.
For simulation of defaults, the initial counts were chosen as they were in September of 2013: \(N^{1,1}(1)= N^{2,1}(1)=30,\) \(N^{3,1}(1)=N^{5,1}(1)=20,\) \(N^{4,1}(1)=25.\) That is, \(\mathcal {N}(1)=125\).
Among all time intervals between 1991 and 2013 that contain sufficiently many credit rating migrations in order to estimate the model parameters, the period from 2000 through 2003 exhibits the greatest probability of default of investment grade debtors for the whole pool of debtors as well as for each of the five industry sectors under consideration. Since the actual iTraxx portfolio contains only debtors of this credit quality, with such choice of a time interval the most intensive flow of defaults should be expected and, thus, it was used for the numerical experiments reported next. Hence, \(t=1\) and \(T=4\) correspond to years 2000 and 2003, respectively.
6 Testing Synchronicity
As a benchmark case allowing for a comparison with known results, first the parameters were estimated assuming that all industry sectors are governed by the same Markovian matrix.
6.1 Common for All Industries Transition Matrix P
A stronger asynchronicity pattern emerged when credit rating migrations in industry sector s were governed by \(P^{(s)}\).
6.2 IndustrySpecific Transition Matrices \(P^{(s)}\)
The data analyzed here indicate that, allowing for industry specific Markovian matrices and asynchronously evolving industries, more realistic models of dependent credit rating transitions can be developed. However, the documented effect of asynchronously moving industries remains just a “natural phenomenon” as long as no implications regarding riskiness of the portfolio in question have been drawn upon it. A quantitative comparison of the flows of defaults generated by the respective coupling schemes is presented in the next section.
7 Synchronicity and Riskiness
Variation of transition probabilities
Upgrading  Downgrading  

\(\chi _{M(s1)\,+\,i}=1\)  \(\; (1q_{i,s})\frac{1p_i^{(s)}}{p_i^{(s)}}100\)  \((1q_{i,s})100\) 
\(\chi _{M(s1)\,+\,i}=0\)  \((1q_{i,s})100\)  \((1q_{i,s})\frac{p_i^{(s)}}{1p_i^{(s)}}100\) 
Variation of a downgrading probability, industryspecific transition matrices
HV\(\setminus \)IS  1  2  3  4  5  

\(i=1\)  \(i=2\)  \(i=1\)  \(i=2\)  \(i=1\)  \(i=2\)  \(i=1\)  \(i=2\)  \(i=1\)  \(i=2\)  
\(\chi _{M(s1)\,+\,i}=1\)  1.53  2.75  0.79  0.83  6.3  3.4  2.38  3.39  0.97  4.07 
\(\chi _{M(s1)\,+\,i}=0\)  24.71  44.91  16.61  17.06  160.37  84.25  50.63  33.58  48.77  108.99 
Variation of a downgrading probability, common for all industries Markovian transition matrix
HV\(\setminus \)IS  1  2  3  4  5  

\(i=1\)  \(i=2\)  \(i=1\)  \(i=2\)  \(i=1\)  \(i=2\)  \(i=1\)  \(i=2\)  \(i=1\)  \(i=2\)  
\(\chi _{M(s1)\,+\,i}=1\)  3.87  2.87  1.74  14.54  6.98  28.30  2.66  5.45  30.07  33.62 
\(\chi _{M(s1)\,+\,i}=0\)  62.51  47.04  28.11  238.67  112.75  463.59  42.97  89.28  485.71  550.07 
A downgrading for an investment grade debtor implies either a migration to the noninvestment grade category or a default, while for a noninvestment grade debtor it means a default. Abbreviations HV and IS stand for hidden variable and industry sector.
Note that there are models of credit rating migrations, where a Markovian matrix, common for all industries, is adjusted according to macroeconomic conditions. See among others Bangia et al. (2002), McNeil and Wendin (2007), Frydman and Schuermann (2008), Korolkiewicz and Elliott (2008), Stefanescu et al. (2009) and Xing et al. (2012).
Since all but two values in Table 3 exceed their counterparts of Table 2, the situation when a unique Markovian matrix governs migrations in all industry sectors should generate a stronger flow of defaults as compared with the case of industryspecific transition matrices. This guess is confirmed next by simulating defaults according to the above coupling schemes and analyzing the corresponding sample distributions.
VaR and CVaR of loss, 1 year
Scheme\(\setminus \)percentile  50  75  90 and \(\textit{CVaR} (0.1)\)  95 and \(\textit{CVaR} (0.05)\) 

AS  0.59/0.59  1.01/1.01  1.47 and 1.71/1.47 and 1.71  1.69 and 2.49/1.71 and 2.47 
AC  0.51/0.50  0.77/0.75  1.35 and 2.13/1.26 and 3.38  1.58 and 2.13/1.49 and 3.38 
SC  0.51/0.50  0.76/0.75  1.33 and 2.42/1.27 and 3.67  1.56 and 2.42/1.50 and 3.67 
VaR and CVaR of loss, 5 years
Scheme\(\setminus \)percentile  0.5  0.75  0.9 and \(\textit{CVaR}(0.1)\)  0.95 and \(\textit{CVaR} (0.05)\) 

AS  3.69/3.69  4.84/4.83  6.02 and 6.96/6.02 and 6.94  6.74 and 7.68/6.73 and 7.67 
AC  2.14/2.13  3.46/3.15  5.46 and 7.88/4.68 and 7.77  7.08 and 9.77/6.73 and 10.43 
SC  2.00/2.12  3.17/3.17  5.88 and 9.73/4.98 and 9.67  8.17 and 12.84/8.21 and 12.31 
VaR and CVaR of loss, 7 years
Scheme\(\setminus \)percentile  0.5  0.75  0.9 and \(\textit{CVaR}(0.1)\)  0.95 and \(\textit{CVaR} (0.05)\) 

AS  6.03/6.03  7.52/7.52  8.87 and 10.31/8.93 and 11.31  9.77 and 11.06/9.82 and 11.06 
AC  3.45/3.42  5.39/4.85  8.16 and 11.90/7.30 and 11.28  10.39 and 13.58/10.11 and 13.32 
SC  3.16/3.44  5.09/5.00  9.16 and 14.03/8.61 and 12.88  11.76 and 17.43/12.18 and 16.13 
In the case of asynchronously moving industries, greater \(\textit{CVaR}\) values (eleven out of twelve) for all three time horizons suggest, that the loss distribution has a heavier tail when migrations in all industries are governed by the same P. The respective percentiles exhibit the same pattern.
Observe that, the larger an entry of Q is, the weaker will be, on the one hand, the dependence of the respective asset from the rest of the portfolio and, on the other hand, the effect of macroeconomic conditions on the respective transition probabilities. Given in Appendix 1 entries corresponding to industryspecific \(P^{(s)}\) exceed their counterparts for a single P. Consequently, as compared with the case of a common P, first, the dependence among assets is weaker and, second, the downgrading probabilities increase less. Both factors together contribute toward higher losses the case of a single P.
Comparing patterns corresponding to asynchronously and synchronously moving industries governed by the same transition matrix, a heavier tail of the loss distribution for all three time horizons can be attributed to the cascade effect discussed in Kaniovski and Pflug (2007) for the setting with synchronously moving industries.
8 Conclusions
Simulations based on S&P’s data show lack of synchronicity among hidden variables shaping credit rating migrations in different industry sectors. This phenomenon indicates, in particular, that macroeconomic conditions affect differently the five industry sectors considered in the paper. It would be interesting to compare for different rating agencies patterns of dependence between hidden variables and, consequently, strength and direction of the corresponding macroeconomic factors.
References
 Allman, E. L., Matias, C., & Rhodes, J. A. (2009). Identifiability of parameters in latent structure with many observed variables. Annals of Statistics, 37(6A), 3099–3132.CrossRefGoogle Scholar
 Bahadur, R. R. (1961). A representation of the joint distribution of responses to \(n\) dichotomous items. In H. Solomon (Ed.), Studies in item analysis and prediction, Stanford University Press, USA, 158–168.Google Scholar
 Bangia, A., Diebold, F. X., Kronimus, A., Schagen, C., & Schuermann, T. (2002). Ratings migration and the business cycle, with applications to credit portfolio stress testing. Journal of Banking and Finance, 26(2/3), 445–474.CrossRefGoogle Scholar
 Boreiko, D. V., Kaniovski, Y. M., & Pflug, G. Ch. (2015). Modeling dependent credit rating transitions—A comparison of coupling schemes and empirical evidence. Central European Journal of Operations Research,. doi: 10.1007/s1010001504156.
 CarreiraPerpiñán, M., & Renals, S. (2000). Practical identifiability of finite mixtures of multivariate Bernoulli distributions. Neural Computation, 12(1), 141–152.CrossRefGoogle Scholar
 ChorośTomczyk, B., Härdle, W., & Okhrin, O. (2013). Valuation of collateralized debt obligations with hierarchical Archimedean copulae. Journal of Empirical Finance, 24(1), 42–62.CrossRefGoogle Scholar
 Frydman, H., & Schuermann, T. (2008). Credit rating dynamics and Markov mixture models. Journal of Banking and Finance, 32(6), 1062–1075.CrossRefGoogle Scholar
 Gupton, G. M., Finger, Ch C, & Bhatia, M. (1997). CreditMetrics—Technical document. New York: J.P. Morgan Inc.Google Scholar
 Hull, J., & White, A. (2004). Valuation of a CDO and an \(n\)th to default CDS without Monte Carlo simulation. Journal of Derivatives, 12(2), 8–23.CrossRefGoogle Scholar
 Kaniovski, Y. M., & Pflug, G Ch. (2007). Risk assessment for credit portfolios: A coupled Markov chain model. Journal of Banking and Finance, 31(8), 2303–2323.CrossRefGoogle Scholar
 Korolkiewicz, M., & Elliott, R. (2008). A hidden Markov model of credit quality. Journal of Economic Dynamics and Control, 32(12), 3807–3819.CrossRefGoogle Scholar
 Li, D. X. (2000). On default correlation: A copula approach. Journal of Fixed Income, 9(4), 43–54.CrossRefGoogle Scholar
 McNeil, A., & Wendin, J. (2007). Bayesian inference for generalized linear mixed models of portfolio credit risk. Journal of Empirical Finance, 14(2), 131–149.CrossRefGoogle Scholar
 Nagpal, K., & Bahar, R. (2001). Measuring default correlation. Risk, 14(3), 129–132.Google Scholar
 Stefanescu, C., Tunary, R., & Turnbull, S. (2009). The credit rating process and estimation of transition probabilities: A Bayesian approach. Journal of Empirical Finance, 16(2), 216–234.Google Scholar
 Wozabal, D., & Hochreiter, R. (2012). A coupled Markov chain approach to credit risk modeling. Journal of Economic Dynamics and Control, 36(3), 403–415.CrossRefGoogle Scholar
 Xing, H., Sun, N., & Chen, Y. (2012). Credit rating dynamics in the presence of unknown structural breaks. Journal of Banking and Finance, 36(1), 78–89.CrossRefGoogle Scholar
Copyright information
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.