Skip to main content
Publications
Journal Article
Management Science 60 (3): 730–752
Ayse Kocabiykoglu, Ioana Popescu, Catalina Stefanescu (2014)
Subject(s)
Product and operations management
Keyword(s)
Revenue management, pricing, coordination, price-sensitive stochastic demand, hierarchical policies, lost sales rate elasticity
The integration of systems for pricing and revenue management must trade off potential revenue gains against significant practical and technical challenges. This dilemma motivates us to investigate the value of coordinating decisions on prices and capacity allocation in a stylized setting. We propose two pairs of sequential policies for making static decisions—on pricing and revenue management—that differ in their degree of integration (hierarchical versus coordinated) and their pricing inputs (deterministic versus stochastic). For a large class of stochastic, price-dependent demand models, we prove that these four heuristics admit tractable solutions satisfying intuitive sensitivity properties. We further evaluate numerically the performance of these policies relative to a fully coordinated model, which is generally intractable. We find it interesting that near-optimal performance is usually achieved by a simple hierarchical policy which sets prices first, based on a non-nested stochastic model, and then uses these prices to optimize nested capacity allocation. This tractable policy largely outperforms its counterpart based on a deterministic pricing model. Jointly optimizing price and allocation decisions for the high-end segment improves performance, but the largest revenue benefits stem from adjusting prices to account for demand risk.
© 2014 INFORMS
Volume
60
Journal Pages
730–752
Journal Article
Management Science 57 (7): 1267–1287
Sudheer Chava, Catalina Stefanescu, Stuart Turnbull (2011)
Subject(s)
Finance, accounting and corporate governance
Keyword(s)
default prediction, expected loss, recovery rate
In this paper we focus on modeling and predicting the loss distribution for credit risky assets such as bonds and loans. We model the probability of default and the recovery rate given default based on shared covariates. We develop a new class of default models that explicitly account for sector specific and regime dependent unobservable heterogeneity in firm characteristics. Based on the analysis of a large default and recovery data set over the horizon 1980-2008, we document that the specification of the default model has a major impact on the predicted loss distribution, while the specification of the recovery model is less important. In particular, we find evidence that industry factors and regime dynamics affect the performance of default models, implying that the appropriate choice of default models for loss prediction will depend on the credit cycle and on portfolio characteristics. Finally, we show that default probabilities and recovery rates predicted out-of-sample are negatively correlated, and that the magnitude of the correlation varies with seniority class, industry, and credit cycle.
© 2011 INFORMS
Volume
57
Journal Pages
1267–1287
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
exchangeable continuous data, maximum likelihood, correlation, accept-reject simulation
This article investigates the Farlie-Gumbel-Morgenstern class of models for exchangeable continuous data. We show how the model specification can account for both individual and cluster level covariates, we derive insights from comparisons with the multivariate normal distribution, and we discuss maximum likelihood inference when a sample of independent clusters of varying sizes is available. We propose a method for maximum likelihood estimation which is an alternative to direct numerical maximization of the likelihood that sometimes exhibits non-convergence problems. We describe an algorithm for generating samples from the exchangeable multivariate Farlie-Gumbel-Morgenstern distribution with any marginals, using the structural properties of the distribution. Finally, we present the results of a simulation study designed to assess the properties of the maximum likelihood estimators, and we illustrate the use of the FGM distributions with the analysis of a small data set from a developmental toxicity study.
With permission of Elsevier
Volume
6
Journal Pages
503–512
Journal Article
Journal of Empirical Finance 16 (2): 216–234
Catalina Stefanescu, Radu Tunaru, Stuart Turnbull (2009)
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
ratings transitions, Bayesian inference, latent factors, Markov Chain, Monte Carlo
JEL Code(s)
G21, G28, G32, C11, C13, C52
The Basel II Accord requires banks to establish rigorous statistical procedures for the estimation and validation of default and ratings transition probabilities. This raises great technical challenges when sufficient default data are not available, as is the case for low default portfolios. We develop a new model that describes the typical internal credit rating process used by banks. The model captures patterns of obligor heterogeneity and ratings migration dependence through unobserved systematic macroeconomic shocks. We describe a Bayesian hierarchical framework for model calibration from historical rating transition data, and show how the predictive performance of the model can be assessed, even with sparse event data. Finally, we analyze a rating transition data set from Standard and Poor's during 1981-2007. Our results have implications for the current Basel II policy debate on the magnitude of default probabilities assigned to low risk assets.
With permission of Elsevier
Volume
16
Journal Pages
216–234
Journal Article
Communications in Statistics - Simulation and Computation 37 (1): 212–221
Catalina Stefanescu, Devan V. Mehrotra (2008)
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
crossover trials, ANCOVA, bootstrap
Volume
37
Journal Pages
212–221
Journal Article
Biometrical Journal 48 (6): 992–1007
Vance W. Berger, Catalina Stefanescu, Yan Yan Zhou (2006)
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
admissibility, binary data, clinical trial, common odds ratio, omnibus test
We consider the problem of testing for independence against the consistent superiority of one treatment over another when the response variable is binary and is compared across two treatments in each of several strata. Specifically, we consider the randomized clinical trial setting. A number of issues arise in this context. First, should tables be combined if there are small or zero margins? Second, should one assume a common odds ratio across strata? Third, if the odds ratios differ across strata, then how does the standard test (based on a common odds ratio) perform? Fourth, are there other analyzes that are more appropriate for handling a situation in which the odds ratios may differ across strata? In addressing these issues we find that the frequently used Cochran-Mantel-Haenszel test may have a poor power profile, despite being optimal when the odds ratios are common. We develop novel tests that are analogous to the Smirnov, modified Smirnov, convex hull, and adaptive tests that have been proposed for ordered categorical data.
© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Volume
48
Journal Pages
992–1007
Journal Article
Technometrics 48 (3): 411–417
Catalina Stefanescu, Bruce W. Turnbull (2006)
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
correlated survival data, Gibbs sampling, load-sharing models, Markov chain Monte Carlo, multivariate lognormal frailty, reliability
Volume
48
Journal Pages
411–417
Journal Article
Biometrical Journal 47 (2): 206–218
Catalina Stefanescu, Bruce W. Turnbull (2005)
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
exchangeable binary data , multivariate binomial probit , Gibbs sampling, hierarchical Bayesian modelling
This paper considers the use of a multivariate binomial probit model for the analysis of correlated exchangeable binary data. The model can naturally accommodate both cluster and individual level covariates, while keeping a fairly flexible intracluster association structure. We discuss Bayesian estimation when a sample of independent clusters of varying sizes are available, and show how Gibbs sampling may be used to derive the posterior densities of parameters. The methodology is illustrated with two examples: the first involves epidemiological data from a study of familial disease aggregation; the second uses teratological data from a developmental toxicity application.
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Volume
47
Journal Pages
206–218
Journal Article
Biometrics 59 (1): 18–24
Catalina Stefanescu, Bruce W. Turnbull (2003)
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
clinical trails, EM algorithm, exchangeable binary data, trend test
Volume
59
Journal Pages
18–24
Journal Article
Environmetrics 12 (7): 659–672
Christina Ahrens, Naomi Altman, George Casella, Malaika Eaton, T. J. Gene Hwang, John Staudenmayer, Catalina Stefanescu (2001)
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
clustering, covariates, confounding, environmental epidemiology
The 1978-1982 New York State TCE / Leukemia dataset is often used to test new cluster detection methodologies. We augment that dataset with demographic covariates from the 1980 census and find evidence that the relation between several of the TCE wastesites and elevated leukemia rates is probably confounded by the population's age and employment characteristics. This demonstrates a problem that is often mentioned, but seldom touched in detail-clustering can be related to covariates not directly related to the risks of interest.
© 2001 John Wiley & Sons, Ltd.
Volume
12
Journal Pages
659–672
Journal Article
Simulation Practice and Theory 6 (7): 657–663
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
Multitype Galton-Watson chain, stochastic process, extinction probability, simulation
We investigate the behaviour of a class of multitype Galton-Watson chains modelling the development of a genetic population. An algorithm for the computer generation of their trajectories is described and details concerning its implementation are given. We analyse through a simulation study the dependence of the extinction time of several populations on the initial gene frequencies and on the individual relative fertilities.
With permission of Elsevier
Volume
6
Journal Pages
657–663
Journal Article
European Association for Theoretical Computer Science Bulletin 66: 139–149
Catalina Stefanescu, Christian Calude, Elena Calude (1998)
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
complementarity, automata
Volume
66
Journal Pages
139–149
Journal Article
Journal of Universal Computer Science 1 (12): 821–827
Subject(s)
Management sciences, decision sciences and quantitative methods
Keyword(s)
Markov process, transition probabilities, memory allocation
Volume
1
Journal Pages
821–827