CC BY-NC-ND
95
ZETA™ METHODOLOGY AND VARIATION IN THE SYSTEMIC RISK OF DEFAULT: ACCOUNTING FOR THE EFFECTS OF TYPE II (FALSE NEGATIVE)
ERRORS VARIATION ON LENDING
Stefano Olgiati, Alessandro Danovi,
Department of Management, Economics and Quantitative Methods, ___________________________________________________University of Bergamo, Italy._
Abstract
This paper argues that a standard Bayesian transformation of the ZETA bankruptcy prediction methodology, introduced by Altman in 1968-1977, allows for an a posteriori update of conditional Type I and II errors due to variation in the systemic likelihood of default. The Bayesian transformation can be used both to condition a loan manager’s prior decision (generally based on Basel II-compliant Internal Rating Based system or Credit Agency’s Rating) and to update such a decision on the basis of any posterior hypothesis (based on actuarial frequentist assumptions of conditional hazard rates) regarding the creditworthiness and the probability of default of an underlying pool of securities.
At the same time - under a Bayesian framework - the ZETA diagnostic test can be conditioned on the new evidence introduced by other tests to increase the total sensitivity of the default prediction models (IRB ratings, TTC ratings, logit, probit, neural) to update the commercial bank’s lending decisions.
A ground-state, static meta-analysis of Altman’s et al. ZETA original article (1977) reveals that the odds of the commercial bank detecting a default after the ZETA score has been introduced (posttest) is 13.2 times more effective than the a priori prediction. Under the same assumptions, the odds of the commercial bank detecting a survival after (post-test) the ZETA score has been introduced is 12.2 times more effective than the a priori.
Key words: Altman’s Z Score, Risk of Default, Bankruptcy prediction, Bayesian Transformation JEL: G32
Introduction and Background
In general modeling correlation and dependence (Li, 2000; Embrechts et al., 2002; Embrechts et al., 2008; Embrechts 2009) of pooled idiosyncratic risks should perform the task of increasing the efficiency of financial markets by transferring some of banks’ credit risks and allow them to free up capital (Donnelly and Embrechts, 2010). However, there is some consensus (Brunnermeier, 2009; Crouhy, 2008; Salmon 2009) that before 2007 the reduction in the standards of idiosyncratic credit risk assessment by lenders prior to risk transfer (Donnelly 2010) - coupled with the worsening of the systemic risk scenario (Stiglitz, 2008) - has led to the well known 2007-2008 financial crisis, with part of the blame falling on the incapacity of the mathematical models (Derman 2013; Lohr, 2009; Donnelly 2010) to update the worst case scenarios on the basis of variation in correlation in the underlying risks pool and the systemic probabilities of default.
The authors of this paper argue that, on the other hand, the Bayesian transformation (Ross, 2002; Gelman et al., 2004) of the ZETA™ scoring methodology introduced by Altman in 1968-1977 allows for a continuous a posteriori update of conditional Type I and II errors due to variation in the systemic likelihood of default (prevalence of disease in epidemiology - Peacock, 2011). The Bayesian transformation can be used both to condition some prior decision - generally based on Basel II-compliant Internal Rating Based system or Credit Agency’s Rating - and to update any posterior hypothesis (based on actuarial frequentist assumptions of conditional hazard rates) regarding the creditworthiness and the probability of default of an underlying pool of securities.
At the same time - under a Bayesian framework - the ZETA™ diagnostic test can be integrated with other tests (Ross Sheldon, 2002; Gelman et al., 2004) to increase the sensitivity of other default prediction models (IRB ratings, TTC ratings, logit, probit, neural) to condition the commercial bank’s lending decisions.
71
Bankruptcy prediction has always been a central theme for banks, which must decide whether or not to approve of credit requests on the basis of estimating the likelihood of default of the borrower at the time of the request and for a reasonable future period. In the 1930s scholars started to develop models using statistics and economic-financial indicators (e.g. Smith, 1930; FitzPatrick, 1931, 1932; Ramser and Foster, 1931; Smith and Winakor, 1935; Wall, 1936). From the 1960s (Tamari, 1966; Beaver, 1966; Altman, 1968, Deakin, 1972, 1977; Joy et al., 1975) to the 1980s discriminant analysis models became the principal approach9.
Among them, probably the most popular is Altman’s Z-score model, formulated in 1968 (Altman, 1968) and composed of a discriminant function based on five variables weighted by coefficients10. The model has been revised several times by its author (Altman, 1983; 2002; etc.) who has constantly updated the parameters and adapted the indices for different populations of companies other than American manufacturers quoted on the Stock Market. The Z’-Score (Altman, 1983) is a variation for private companies. The Z” Score (Altman, Hartzell and Peck, 1995 and Altman & Hotchkiss, 2006, p. 314) was introduced for the non-manufacturing sector or companies operating in developing countries (the original study investigated a sample of Mexican companies) and was recently tested for Italy (Altman, Danovi, Falini 2013). Another adaptation was the introduction of the Z-Metrics System (Altman, Rijken et. al., 2010) that refines the original model, including both market equity levels and volatility, as well as macro-economic variables.
Research Questions
In medical statistics the Bayesian approach to prior hypotheses testing and updating when new evidence has been introduced is uncontroversial and represents the contemporary epidemiological standard for clinical and pharmacological research (Peackock 2011). In particular, Bayesian transformations are widely used when there is a rapid variation in the prevalence and incidence in the underlying disease pool during an epidemic or a pandemic (Ashby, 2006; Bland 1998) and the physician needs to update the probability of Type I and II diagnostic errors.
The hypothesis of the authors of transferring the epidemiological experience to bankruptcy classification rests on the two following assumptions:
1. the credit manager will utilize the ZETA methodology to condition some a priori expectation (probability distribution) based on some Basel II-compliant Internal Rating Based system or Credit Agency’s Rating regarding the probability of default of the corporation he is evaluating. However, the conditional probability that the credit manager will grant or withhold credit on the basis of the further evidence provided by the ZETA analysis is not the same as the underlying true probability of bankruptcy of the corporation (Ross Sheldon, 2002; Gelman et al., 2004), and neither it is the same as the probability of the corporation getting a negative ZETA if they are effectively going to go bankrupt (Ross Sheldon, 2002; Gelman et al., 2004);
2. variation in the systemic risk of default P(D) of the credit pool should affect the credit manager’s belief (Ross Sheldon, 2002; Gelman et al., 2004) in the likelihood of default of the corporation he is evaluating, even if the ZETA methodology discriminant coefficients are based on idiosyncratic risks and are not updated continuously.
We performed a ground-state, static meta-analysis of Altman’s ZETA original article (1977) - to which we refer in full - and transformed the variables under a standard Bayesian transformation.
In this sense, these are the research questions:
1. we would expect the commercial bank’s conditional estimates of default P(D) given that
9. Univariate and multivariate statistical analysis are still in use (Edmister, 1972; Blum, 1974; Elam, 1975; Libby, 1975; Taffler, 1976, 1982; Altman et. al., 1977, 1993; Wilcox, 1976; Argenti, 1976; Appetiti, 1984; Lawrence and Bear, 1986; Aziz, Emanuel and Lawson, 1988; Baldwin and Glezen, 1992; Flagg, Giroux and Wiggins, 1991; Bijnen and Wijn, 1994; Kern and Rudolph, 2001; Altman, Rijken, et. al., 2010 Kruchynenko, 2012). For a comprehensive overview see Bellovary, Giacomino e Akers (2007) that summarizes 165 works from the 1930s to 2007.
10. The first application of the model involved a group of 66 American manufacturing companies (33 healthy and 33 bankrupt), listed on the Stock Exchange and showed that companies with a Z Score of less than 1.81 were highly risky and likely to go bankrupt; companies with a score more than 2.99 were healthy and scores between 1.81 and 2.99 were in a grey area with uncertain results). The model was extremely accurate since the percentage of correct predictions was about 95% and it received many positive reactions and only a few criticisms.
72
Z is negative (loan rejected) P(D|Z<2.675) to be different from the discriminant’s analysis probabilities of being negative given actual default P(Z<2.675|D);
P(D)
2. the pre-test effectiveness (odds ratio) PD) of detecting a default increases by the post-test
ratio P(z 1 D) . ratio p(z | Dc) ;
3. reveal a commercial bank’s lower tolerance for conditional Type I Errors and a higher tolerance for conditional Type II Errors. i.e. P(D|Zc)<< P(Dc|Z);
4. verify that the Bayesian framework permits expansion and integration of the contribution P(Dj||Z) that the ZETA Analysis gives to any prior hypotheses Dj (TTC ratings, neural networks, logit, probit) about the probability of default of the borrower;
5. verify that ZETA Analysis can be continuously updated on the basis of data relative to the systemic hazard rate of the underlying risks pool P(D);
6. confirm the postulate that the ZETA test causes, in epidemiological jargon and on a macroeconomic level, a selection of the fittest corporations (patients) amongst the fittest (cream skimming) (Levaggi, 2003) by the commercial bank;
7. confirm that the pre-test and post-test efficiency comparison ECzeta=.00243 is invariant under the Bayesian transformation.
All datasets, variables and parameters utilized in the Meta-Analysis are from Altman EI, Halde-man R.G., Narayanan P. (1977), “ZETA™ Analysis”, Journal of Banking and Finance; 1: 29-54. The version utilized can be accessed and freely downloaded from E. Altman’s website at the New York University Stern School of Business at http://people.stern.nyu.edu/ealtman/papers.html (last accessed April 2013).
The Bayesian notation utilized is coherent with Berkley University’s textbook: Ross Sheldon (2002), “Probability”, 6th Ed., Prentice Hall (NJ), USA; 3: 64-121.
The epidemiological approach to Bayesian statistics is coherent with Oxford University’s textbook: Peacock JL, Peacock PJ. (2011), “Medical Statistics”, Oxford University Press, UK; 7: 234-5 and 14: 478-97.
Methods and Findings
We partitioned Altman et al. ZETA™ Analysis (1977) sample space ^=P(D)+P(Z)+P(DcnZc)=1 on the basis of mixed probabilities P(ZflD)=.018, P(DflZc)=.002, P(DcflZ)=.069 and P(DcflZc)=.911 and we conditioned the a priori probabilities of default P(D)=.02 on the a posteriori evidence of the ZETA tests calculated by Altman (1977).
We utilized Bayes’ Formula to condition the probability of default P(D|*) and of survival P(Dc|*) on P(*|Z) given the a priori probability of default P(D)=.02, of survival [P(Dc)=1-P(D)]=.98, the a posteriori experimental accuracy of the ZETA test P(Z|D)=.924 and P(Zc|Dc)=.930, and the experimental Type II P(Z|Dc)=.070 and Type I P(Zc|D)=.076 Errors.
In formal terms:
P(D| Z) = PZ^PD
and
P(Z | D)P(D) + P(Z | Dc )[1 - P(D)]
P(Dc | Zc) = -
P(Zc|Dc )|~1- P(D)]
’ P(Zc |Dc)[1-P(D)] +P(Zc |D)P{D) .
We then applied the results to the test utilized by Altman in his original paper (Altman et al. 1977, page 43) to compare the efficiency of the ZETA™ methodology with alternative strategies, which will be invariant after the Bayesian transformation if:
i, («„/N +q (MJ N - P (Zc > («!Zc+P (Z > (D lZ U
73
and therefore:
EC = EC
ZETA ZETA-BAYES
.00243
where qpq2 are the prior probabilities of bankrupt (qt) or non-bankrupt (q2); C: (Type I Error) is the cost of an accepted loan that defaults, CII is the cost of a rejected loan that would have resulted in a successful payoff; M12,M21 are the observed Type I and II errors (misses) respectively; and NPN2 are the number of observations in the bankrupt (N1) and non-bankrupt (N2) groups (Altman et al., 1977).
All findings are summarized in Exhibit A.
The conditional probability of the commercial bank detecting a default when the corporation will actually default after the ZETA test has been introduced is P(D|Z)=21.2% (circa 2 out of 10), with a Type II Error P(Dc|Z)=78.8% (circa 8 out of 10).
The conditional probability of the commercial bank detecting a healthy corporation when the corporation will actually not go bankrupt (survive) after the ZETA test has been introduced is P(Dc|Zc)=99.8% (circa 10 out of 10), with a Type I Error P(D|Zc)=0.2% (circa 0 out of 10).
The odds of the commercial bank detecting a default after (post-test) the ZETA score has been
P (pIZ )=.269 P (ZID)
= 13.2
P(D)
= .02
introduced P ( 1Z) is p(z I p) '
times more effective than the a priori P(Dc)
The odds of the commercial bank detecting a default after (post-test) the ZETA score has been in-
P (d IZ)
= 599.6
P (z I d)
troduced P(D 1Z) is P(Z1D )
The Efficiency Comparison Test ECzeta: was expected, in fact:
12 2 P(D)
12.2 —= 49.0
times more effective than the a priori p(pc)
=.00243 is invariant after the Bayesian transformation as
P (zcD)pi + P (ZDC C = P (DZc C + P (pcZ C = .00243
Exhibit A: Bayesian reduced sample space partitioning of Altman’s ZETA Analysis (1977)
P(D) Prior hypothesis - Probability of default P(D) 0.020 2.0%
P(Dc) Prior hypothesis - Probability of survival [1-P(D)] 0.980 98.0%
P(Z) Probability of negative Z<2.675 0.087 8.7%
P(Zc) Probability of positive Z>2.675 0.913 91.3%
P(Z|D) Probability of negative Z<2.675 given default 0.924 92.4%
P(Zc|Dc) Probability of positive Z>2.675 given survival 0.930 93.0%
P(Zc|D) Type I - Probability of positive Z>2.675 given default 0.076 7.6%
P(Z|Dc) Type II - Probability of negative Z<2.675 given survival 0.070 7.0%
P(D|Z) Probability of default given that Z<2.675 is negative 0.212 21.2%
P(Dc|Zc) Probability of survival given that Z>2.675 is positive 0.998 99.8%
P(D|Zc) Type I - Probability of default given Z>2.675 is positive 0.002 0.2%
P(Dc|Z) Type II - Probability of survival given Z<2.675 is negative 0.788 78.8%
P(ZD) negative Z<2.675 and default 0.018 1.8%
P(DZ) negative Z<2.675 and default 0.018 1.8%
P(ZDc) negative Z<2.675 and survival 0.069 6.9%
P(DcZ) negative Z<2.675 and survival 0.069 6.9%
P(ZcDc) positive Z>2.675 and survival 0.911 91.1%
P(DcZc) positive Z>2.675 and survival 0.911 91.1%
P(DZc) positive Z>2.675 and default 0.002 0.2%
74
P(ZcD) positive Z>2.675 and default 0.002 0.2%
S Space of events (Summation) 1.000 100.0%
P(D) / P(Dc) Odds ratio of default 0.020
PZ|D) / P(Z|Dc) Likelihood 13.2
P(D|Z) / P(Dc|Z) New odds ratio of default after the Z-test has been introduced 0.269
P(Dc) / P(D) Odds ratio of survival 49.000
P(Zc|Dc) / P(Zc|D) Likelihood 12.2
P(Dc|Zc) / P(D|Zc) New odds ratio of survival after the Z-test has been introduced 599.6
Source: Authors’ elaborations on Altman’s ZETA Analysis (1977)
Medicine students and practitioners are usually very surprised by the fact that any reliable diagnostic test, with an accuracy of 92.4%, Type I errors of 7.6% and Type II Errors of 7.0%, is capable of diagnosing a patient affected by a disease when the patient actually has that disease only 2 out of 10 times (Ross 2002).
They are normally subject to the instinctive - but sometimes severely misleading - a priori belief in the perfect correlation between the outcome of the diagnosis (Z) and the disease itself (D), and in the a posteriori belief that the probability of diagnosing the disease D given a positive test Z is a linear function of utilizing a test Z which is positive given that the disease is present (Peacock 2011). The reason is that Bayes’ rule states that the probability of the event default D is a weighted average of the conditional probability of P(D|*) given that Z=ZETA<2.675 has occurred and the conditional probability of D given that Z has not occurred (Zc=ZETA>2.675), each conditional probability is given as much weight as the event P(D) or P(Dc) on which it is conditioned to occur (Bayes, 1763). In addition, the Bayesian transformation of Type II errors from P(Z|Dc)=7% to P(Dc|Z)=78.8% is so steep at low P(D)s (2% in the 1977 ZETA Model) that, at such low P(D)s, increases in Type II Errors of the ZETA test P(Z|Dc) (>7%) (Exhibit B) implies a growing number of creditworthy corporations (> 8 out of 10) whose requests for loans are unexplainably rejected.
ZETA Analysis, after the Bayesian transformation consciously or unconsciously performed by the commercial bank’s credit manager selects, from a macroeconomic point of view and according to epidemiological jargon, the fittest amongst the fittest (cream skimming).
In this sense it appears to be also a very effective managerial (clinical) tool to diagnose the conditional health of a corporation, with a P(Dc|Zc)=99.8% (almost 10 out of 10).
Bayesian model averaging (BMA) in forecasting has been analyzed and described in detail by Stock and Watson (2004). When idiosyncratic credit default prediction models are combined, the forecast is a weighted average of the individual model outcomes, where the weights depend on the accuracy of the individual forecasts. In BMA the weights are computed as posterior probabilities that the model is correct. In addition, the individual forecasts in BMA are model-based and the posterior means of the variable are forecast, conditioned on the selected model. Thus BMA extends forecast combining to a fully Bayesian setting, where the forecasts themselves are Bayes forecasts, given the model (and some parametric priors) (Stock and Watson, 2004).
In other words, if the commercial bank utilizes several scoring and rating systems and is e.g. P(Dj) =60% confident that the borrower will default, the ZETA Analysis will condition the a priori probabilities of default P(Dj|*) of several other tests on Z so that:
, 4 p(z | D.Wd.)
P(d. | Z)= ^ = 95%
±p(z | д)>(д)
7=1
Exhibit B: Relationship between decisor’s conditional P(D|Z) and ZETA Analysis P(Z|D) at P(D)=02 and Type II Error P(Z|Dc)=07
75
E. Altman’s ZETA (1977) bankruptcy classification model is based on the sound and empirically tested scientific principles of discriminant analysis which comply with the ROBUST (Reporting Of Bayes Used in clinical STudies) criteria of conditional Bayesian analysis (Sung 2005; Spiegelhalter et al., 1999a; Spiegelhalter et al., 1999b; Spiegelhalter et al., 2000).
In particular:
• the prior distribution is specified - normal distribution;
• the prior distribution is justified - historical bankruptcy data;
• Type I and II errors are specified;
• the statistical model is specified - discriminant analysis;
• the statistical technique is specified;
• the central tendency tests have been performed;
• the standard deviation and confidence intervals have been specified.
Conclusions
The authors of this paper conclude that a standard Bayesian transformation of the ZETA bankruptcy prediction methodology introduced by Altman in 1968-1977 allows for a continuous a posteriori update of conditional Type I and II errors due to variation in the systemic likelihood of default. The Bayesian transformation can be used both to condition the loan manager’s prior decision (generally based on Basel II-compliant Internal Rating Based system or Credit Agency’s Rating) and to update such a decision on the basis of any posterior hypothesis (based on actuarial frequentist assumptions of conditional hazard rates) regarding the creditworthiness and the probability of default of an underlying pool of securities.
At the same time - under a Bayesian framework - the ZETA diagnostic test can be conditioned on the new evidence introduced by other tests to increase the total sensitivity of the default prediction models (IRB ratings, TTC ratings, logit, probit, neural) to update the commercial bank’s lending decisions.
A ground-state, static meta-analysis of Altman’s et al. ZETA original article (1977) reveals that the odds of the commercial bank detecting a default after the ZETA score has been introduced (post-
76
test) is 13.2 times more effective than the a priori prediction. Under the same assumptions, the odds of the commercial bank detecting a survival after (post-test) the ZETA score has been introduced is 12.2 times more effective than the a priori. Integration of the ZETA model with other default prediction models reaches a credibility interval of CI > 95% when the updated likelihood of default is equal to 60%. As expected, the Efficiency Comparison Test ECzeta=.00243 is invariant under the Bayesian transformation.
Discussion and further survey of economic literature on Bayesian modeling
In standard Bayesian analysis, the parameters of a given model are treated as random variables, distributed according to a prior distribution. In Bayesian Model Averaging (BMA), the binary variables indicating whether a given model is true are also treated as random variables and distributed according to some prior distribution (Stock and Watson, 2004).
If the utilization of Bayesian average modeling is undisputed in epidemiology (Peackock 2011; Ashby, 2006; Bland 1998) with a wider use advocated (Lilford, 1996), the applications of BMA to economic forecasting have been quite recent (Stock and Watson, 2004) and have been mostly utilized to predict stock market growth and fluctuations where a continuous update of prior predictors is necessary.
Min and Zellner (1990) have used Bayesian and non-Bayesian methods for combining models and forecasts with applications to forecasting international growth; Avramov (2002) has used BMA to analyze stock return predictability and model uncertainty; Cremers (2002) - to analyze stock return predictability and Wright (2004; 2008) - to forecast inflation by Bayesian model averaging and exchange rate forecasts.
Macroeconomic aggregates have been analyzed by Koop and Potter (2003) who focused on forecasting GDP and the change of inflation.
In general, Bayesian statistics, as opposed to frequentist statistics, have shown some limits in advanced applications, in particular in the substitution of credible intervals (posterior interval) with confidence intervals (CIs) (Gelman et al., 2004; Spiegelhalter et al., 2004), and medical statistics require in general that both techniques be utilized (Deeks at al., 2004; Jones et al., 2007; Kovacs et al., 2001).
Further integration with the wide experiences collected by epidemiologists in analyzing extremely varying disease pools and the associated diagnostics can be useful in developing useful testing for the prediction of bankruptcy (strep throats!).
References
1. Altman E.I., Haldeman R.G., Narayanan P. (1977). ZETA™ Analysis. Journal of Banking and Finance, vol. 1, pp. 29-54.
2. Altman E.I. (1968). Financial Ratios. Discriminant Analysis and the Prediction of Corporate Bankruptcy. The Journal of Finance, pp. 589-609.
3. Altman E.I. (1970). Ratio Analysis and the Prediction of Firm Failure: a Reply. The Journal of Finance, vol 25, no. 25, pp. 1169-1172.
4. Altman E.I., Haldeman R., Narayanan P. (1977). ZETA Analysis: a New Model to Identify Bankruptcy Risk of Corporations. Journal of Banking and Finance, vol. 10. pp. 29-54. Altman E.I. (1983). Corporate Financial Distress. Wiley Interscience. New York.
5. Altman E.I., La Fleur J.K. (1985). I modelli di previsione delle insolvenze: le loro applicazioni alla gestione d’impresa. Finanza, Marketing, Produzione, no. 4, December, pp. 77-93.
6. Altman E.I. (1993). Corporate Financial Distress and Bankruptcy: a Complete Guide to Predicting and Avoiding Distress and Profiting from Bankruptcy. Wiley. New York.
7. Altman E.I., Hartzell J., Peck M. (1995). Emerging Markets Corporate Bonds: a Scoring System. Salomon Brothers Inc. New York. and in Levich R., Mei J.P. The Future of Emerging Market Flaws. Kluwer Publishing. and revisited in Altman E.I., Hotchkiss E. (2006).
77
Corporate Financial Distress & Bankruptcy. J. Wiley & Sons. New York.
8. Altman E.I. (2002). “Revisiting Credit Scoring Models in a Basel 2 Environment”. Ong M. Credit Rating: Methodologies. Rationale and Default Risk. London Risk Book.
9. Altman E.I., Hotchkiss E. (2005). Corporate Financial Distress & Bankruptcy. 3rd edition. J. Wiley & Sons. Hoboken. New Jersey.
10. Altman E.I., Resti A., Sironi A. (2005). Recovery Risk: the Next Challenge in Credit Risk Management. Risk Books. London.
11. Altman E, Rijken H. (2010), The Z-MetricsTMMethodology For Estimating Company Credit Ratings And Default Risk Probabilities, The RiskMetrics Group, Inc., New York.
12. Altman E.I., Danovi A., Falini A. (2013), “Bankruptcy Prediction; Z’ and Z’’ score application to Italian companies subject to Extraordinary Administration”, Journal of Applied Finance, no. 1.
13. Appetiti S. (1984). “L’Utilizzo dell’Analisi Discriminatoria per la Previsione delle Insolvenze: Ipotesi e Test per un’Analisi Dinamica”. Servizio Studi della Banca d’ltalia. Temi di Discussione. Roma.
14. Argenti J. (1976). Corporate Collapse: the Causes and Symptoms. McGraw-Hill. Milano.
15. Argenti J. (1983). Predicting Corporate Failure. Institute of Chartered Accountants in England & Wales. Milton Keynes.
16. Ashby D. (2006), “Bayesian statistics in medicine: a 25 year review”, StatMed. vol. 25, pp. 3589-3631.
17. Avramov, D. (2002), “Stock return predictability and model uncertainty”, Journal of Financial Economics, vol. 64, pp. 423-258.
18. Aziz A., Emanuel D.C., Lawson G.H. (1988). “Bankruptcy Prediction - An Investigation of Cash Flow Based Models”. Journal of Management Studies, vol. 25. no. 5. pp. 419-437.
19. Balwind J., Glezen G. (1992). “Bankruptcy Prediction Using Quarterly Financial Statement Data”. Journal of Accounting, Auditing & Finance, vol. 3. no. 7. pp. 269-285.
20. Bayes T. (1763), “An essay towards solving a problem in the doctrine of chances”, Philos Trans Roy Soc, vol. 53, pp. 370-418.
21. Beaver W.H. (1966). “Financial Ratios as Predictors of Failure. Empirical Research in Accounting: Selected Studies”. Journal of Accounting Research. Supp., pp. 71-111.
22. Bellovary Jodi L., Akers Michael D., Giacomino Don E. (2007). “A Review of Bankruptcy Prediction Studies: 1930 to Present”. Journal of Financial Education, vol. 33. Winter 2007. pp.1-43.
23. Bijnen E.J.. Wijn M.F.C.M. (1994). “Corporate Prediction Models. Ratios or Regression Analysis?” Vol. 658 of FEW. Tilburg University. Faculty of Economics and Business Administration.
24. Bland J.M., Altman D.G. (1998), “Statistics notes: Bayesians and frequentists”, British Medical Journal (BMJ), vol. 317, pp. 1151-1160.
25. Blum M. (1974). “Failing Company Discriminant Analysis”. Journal of Accounting Research, vol. 12, no. 1, pp. 1-25.
26. Brunnermeier M.K. (2009) “Deciphering the liquidity and credit crunch 2007-2008”, Journal of Economic Perspectives, vol. 23, no. 1, pp. 77-100.
27. Cremers K.J.M. (2002), “Stock return predictability: a Bayesian model selection perspective”, The Review of Financial Studies, vol. 15, pp. 1223-1249.
28. Crouhy M. G., Jarrow R. A. and Turnbull, S. M. (2008) “The subprime credit crisis of 07”, Journal of Derivatives, vol. 16, no. 4, pp. 81-110.
29. Deakin E.B. (1972). “A Discriminant Analysis of Predictors of Business Failure”. in Journal
78
of Accounting Research, vol. 10, no. 1, pp. 167-179.
30. Deakin E.B. (1977). Business Failure Prediction: an Empirical Analysis. In Financial Crisis, Institutions and Markets in a Fragile Environment, edited by Altman E.I. and Sametz A.W. John Wiley and Sons. New York, pp.72-98.
31. Deeks JJ, Altman DG (2004), “Diagnostic tests 4: likelihood ratios”, British Medical Journal (BMJ), vol. 329, pp. 168-169.
32. Derman M (2013) in Garud I, Haugh M, Derman M, “Lectures on the implementation difficulties with mean-variance: beyond variance”, Columbia University, Industrial Engineering and Operations Research, unpublished: 4-12.
33. Donnelly C., Embrechts P. (2010), “The devil is in the tails: actuarial mathematics and the subprime mortgage crisis”, ASTIN Colloquium, Helsinki.
34. Edmister R.O. (1972). “An Empirical Test of Financial Ratio Analysis for Small Business Failure Prediction”. in Journal of Financial and Quantitative Analysis, vol. 7, no. 2, pp. 1477-1493.
35. Elam R. (1975). “The Effect of Lease Data on the Predictive Ability of Financial Ratios”. The Accounting Review, vol. 50, no. 1, pp. 25-43.
36. Embrechts P. (2009), “Copulas: a personal view”, Journal of Risk and Insurance, vol. 76, no. 3, pp. 639-650.
37. Embrechts P, Kluppelberg C. and Mikosch T. (2008), Modeling Extremal Events for Insurance and Finance, Springer Verlag, Germany
38. Embrechts P., McNeil A.J. and Straumann D. (2002). “Correlation and dependence in risk management: properties and pitfalls”, in M. Dempster, editor, Risk Management: Value at Risk and Beyond, pages 176-223. Cambridge University Press. First appeared in 1998 as a RiskLab ETH Zurich report.
39. FitzPatrick PJ. (1931). Symptoms of Industrial Failures as Revealed by an Analysis of the Financial Statements of Failed Companies, 1920-1929. The Catholic University of America. Washington.
40. FitzPatrick PJ. (1932). “A Comparison of Ratios of Successful Industrial Enterprises with Those of Failed Companies”. The Certified Public Accountant. October. pp. 598-605. November. pp. 656-662. December. pp. 727-731.
41. Flagg J., Giroux G., Wiggins C. (1991). “Predicting Corporate Bankruptcy Using Failing Firms”. in Review of Financial Economics, vol. 1, no. 4, pp. 67-78.
42. Gelman A., Carlin J.B., Stern Hal S., Rubin D.B. (2004), Bayesian data analysis, 2nd Ed., Chapman & Hall, Boca Raton (FL), USA.
43. Hull J. White A. (2004), “Valuation of a CDO and an nth to Default CDS Without Monte Carlo Simulation”, Journal of Derivatives, vol. 12, no. 2, Winter 2004, p. 12.
44. Hull J. (2009), “Options, Futures and Derivatives”, Prentice Hall, vol. 22, pp. 504-508.
45. Jones R., Latinovic R., Charlton J., Gulliford M.C. (2007), “Alarm symptoms in early diagnosis of cancer in primary care: cohort study using General Practice Research Database”, British Medical Journal (BMJ)vol. 334, p. 1040.
46. Joy M.O., Tollefson J.O. (1975). “On the Financial Applications of Discriminant Analysis”. Journal of Financial and Quantitative Analysis. vol. 10, pp. 723-739.
47. Koop G. and Potter S. (2003), “Forecasting in large macroeconomic panels using Bayesian model averaging”, Working Paper 04/16, University of Leicester.
48. Kern M.. Rudolph B. (2001). “Comparative Analysis of Alternative Credit Risk Models: an Application on German Middle Market Loan Portfolios”, Center for Financial Studies Working Paper, vol. 3.
79
49. Kovacs M.J. et al (2001), “A comparison of three rapid D-dimer methods for the diagnosis of venous thromboembolism”, Br JHaematol, vol. 115, pp. 140-144.
50. Kruchynenko I. (2012). Financial Risk and Models of its Measurement Altman’s Z-Score review. LAP Lambert Academic Publishing.
51. Lawrence E.C.. Bear R.M. (1986). “Corporate Bankruptcy Prediction and the Impact of Leases”. Journal of Business, Finance and Accounting, vol. 13, no. 4, pp. 571-585.
52. Levaggi R. et al. (2003), “Horizontal and vertical cream skimming in the health care market”, Diritti, Regole, Mercato, XV Conferenza SIEP, Pavia, Universita, 3 - 4 ottobre 2003.
53. Li D. (2000), “On default correlation: a copula function approach”, Journal of Fixed Income, vol. 9, no. 4, pp. 43-54.
54. Libby R. (1975). “Accounting Ratios and the Prediction of Failure: Some Behavioral Evidence”. Journal of Accounting Research, vol. 13, no. 1, pp. 150-161.
55. Lilford R.J., Braunholtz D. (1996), “The statistical basis of public policy: a paradigm shift is overdue”, British Medical Journal (BMJ); 313: 603-7
56. Lohr, S. (2009) “Wall Street’s math wizards forgot a few variables”, New York Times, September 12 2009.
57. Min C. and A. Zellner (1993), “Bayesian and non-Bayesian methods for combining models and forecasts with applications to forecasting international growth rates”, Journal of Econometrics, vol. 56, pp. 89-118.
58. Ramser J.R.. Foster L.O. (1931). “A Demonstration of Ratio Analysis”. Bureau of Business Research Bulletin, no. 40, University of Illinois.
59. Smith F.R. (1930). “A Test Analysis of Unsuccessful Industry Companies”. Bureau of Business Research, no. 31, University of Illinois.
60. Smith F.R.. Winakor A.H. (1935), “Changes in Financial Structure of Unsuccessful Corporations”. Bureau of Business Research, no. 31, University of Illinois.
61. Peacock J.L., Peacock PJ. (2011), “Medical Statistics”, Oxford University Press, UK, vol. 7, pp. 234-235 and vol. 14, pp. 478-497.
62. Ross S. (2002), “Probability”, 6th Ed., Prentice Hall (NJ), USA, vol. 3, pp. 64-121.
63. Salmon F. (2009). Recipe for disaster: the formula that killed Wall Street. February 23, Wired Magazine.
64. Spiegelhalter D.J. et al (1999a), “Methods in health service research: an introduction to bayesian methods in health technology assessment”, British Medical Journal (BMJ), vol. 319, pp. 508-512.
65. Spiegelhalter D.J. et al (1999b), Bayesian approaches to clinical trials and health-care evaluation, Wiley, hichester, West Sussex (UK).
66. Spiegelhalter D.J. et al (2000), “Bayesian methods in health technology assessment: a review”, Health TechnolAssess, vol. 4, p. 38.
67. Sung L. et al (2005), “Seven items were identified for inclusion when reporting a Bayesian analysis of a clinical study”, J Clin Epidemiol, vol. 58, pp. 261-268.
68. Stiglitz J. E. (2008). It doesn’t take Nostradamus. The Economists’ Voice, vol. 5, no. 8, article no. 1.
69. Stock J.H. and Watson M.W. (2004), “Forecasting with many predictors”, The Handbook of Economic Forecasting, vol. 5, pp. 25-39. available at NYU Stern http://www.stern.nyu.edu/ experience-stern/about/departments-centers-initiatives/academic-departments/economics/ research/working-papers/2003/ (accessed April 2013).
70. Tamari M. (1966). “Financial ratios as a means of forecasting bankruptcy”. Management International Review. vol. 6, no. 4.
80
71. Taffler R.J. (1976). Finding those Firms in Danger. City University Business School. London. Working Paper No. 3.
72. Taffler R.J. (1982). “Forecasting Company Failure in the UK Using Discriminant Analysis and Financial Ratio Data”. Journal of the Royal Statistical Society. Series A, vol. 145, no. 3, pp. 342-358.
73. Wall A. (1936). How to evaluate financial statements. Harper. New York.
74. Whetten D. (1987). “Organizational growth and decline processes”. Annual Review of Sociology, vol. 13, pp. 335-358.
75. Wilcox J.W. (1976). “The Gambler’s Ruin Approach to Business Risk”. Sloan Management Review, no. 1, fall, pp. 33-46.
76. Wright J.H. (2004), “Forecasting inflation by Bayesian model averaging”, manuscript, Board of Governors of the Federal Reserve System in Stock JH and Watson MW (2004), “Forecasting with many predictors”, The Handbook of Economic Forecasting, vol. 5, pp. 2539 available at NYU Stern http://www.stern.nyu.edu/experience-stern/about/departments-centers-initiatives/academic-departments/economics/research/working-papers/2003/ (accessed April 2013)
77. Wright J.H. (2008), “Bayesian Model Averaging and exchange rate forecasts”, Journal of Econometrics, vol. 146, no. 2, pp. 329-341.
81
