New Economics Papers
on Computational Economics
Issue of 2008‒02‒09
ten papers chosen by

  1. A Genetic Algorithm for the Multi-Mode Resource-Constrained Project Scheduling Problem By V. VAN PETEGHEM; M. VANHOUCKE
  2. A genetic algorithm approach for the single machine scheduling problem with linear earliness and quadratic tardiness penalties By Jorge M. S. Valente; José Fernando Gonçalves
  3. Branching Strategies in a Branch-and-Price Approach for a Multiple Objective Nurse Scheduling Problem By B. MAENHOUT; M. VANHOUCKE
  4. Estimating probabilities of default with support vector machines By Härdle, Wolfgang; Moro, Rouslan A.; Schäfer, Dorothea
  5. Solving, Estimating and Selecting Nonlinear Dynamic Models without the Curse of Dimensionality By Viktor Winschel; Markus Krätzig
  6. Can Good Events Lead to Bad Outcomes? Endogenous Banking Crises and Fiscal Policy Responses By Andrew Feltenstein; Céline Rochon
  9. Classification Using Association Rules By Dass Rajanish
  10. Trade Liberalization in Latin America and Eastern Europe: The Cases of Ecuador and Slovenia By Sang-Wook Stanley Cho; Julian P. Diaz

    Abstract: In this paper we present a genetic algorithm for the multi-mode resource-constrained project scheduling problem (MRCPSP), in which multiple execution modes are available for each of the activities of the project. In contrast to a conventional genetic algorithm, we apply a bi-population genetic algorithm, which makes use of two seperate populations. We extend the serial schedule generation scheme by introducing a mode optimization procedure.We present detailed comparative computational results, which reveals that our procedure is among the most competitive algorithms for the MRCPSP.
    Keywords: project scheduling, genetic algorithm, multi-mode RCPSP
    Date: 2008–01
  2. By: Jorge M. S. Valente (LIAAD, Faculdade de Economia, Universidade do Porto, Portugal); José Fernando Gonçalves (LIAAD, Faculdade de Economia, Universidade do Porto, Portugal)
    Abstract: In this paper, we consider the single machine scheduling problem with linear earliness and quadratic tardiness costs, and no machine idle time. We propose a genetic approach based on a random key alphabet. Several genetic algorithms based on this approach are presented. These versions differ on the generation of the initial population, as well as on the use of local search. The proposed procedures are compared with the best existing heuristic, as well as with optimal solutions for the smaller instance sizes. The computational results show that the performance of the proposed genetic approach is improved by the addition of a local search procedure, as well as by the insertion of simple heuristic solutions in the initial population. Indeed, the genetic versions that include either or both of these features not only provide significantly better results, but are also much faster. The genetic versions that use local search are clearly superior to the best existing heuristic, and the improvement in performance increases with both the size and difficulty of the instances. These procedures are also quite close to the optimum, and provided an optimal solution for most of the test instances.
    Keywords: scheduling, single machine, linear earliness, quadratic tardiness, genetic algorithms
    Date: 2008–01
    Abstract: The efficient management of nursing personnel is of critical importance in a hospital’s environment comprising a vast share of the hospital’s operational costs. The nurse scheduling process affects highly the nurses’ working conditions, which are strongly related to the provided quality of care. In this paper, we consider the rostering over a mid-term period that involves the construction of duty timetables for a set of heterogeneous nurses. In scheduling nursing personnel, the head nurse is typically confronted with various (conflicting) goals complying with different priority levels, which represent the hospital’s policies and the nurses’ preferences. In constructing a nurse roster, nurses need to be assigned to shifts in order to maximize the quality of the constructed timetable satisfying the case-specific time related constraints imposed on the individual nurses’ schedules. Personnel rostering in healthcare institutions is a highly constrained and difficult problem to solve and is known to be NP-hard. In this paper, we present an exact branch-and-price algorithm for solving the nurse scheduling problem incorporating multiple objectives and discuss different branching and pruning strategies. Detailed computational results are presented comparing the proposed branching strategies and indicating the beneficial effect of various principles encouraging computational efficiency.
    Keywords: Nurse Scheduling, Branch-and-Price, Branching Strategies
    Date: 2008–01
  4. By: Härdle, Wolfgang; Moro, Rouslan A.; Schäfer, Dorothea
    Abstract: This paper proposes a rating methodology that is based on a non-linear classification method, the support vector machine, and a non-parametric technique for mapping rating scores into probabilities of default. We give an introduction to underlying statistical models and represent the results of testing our approach on Deutsche Bundesbank data. In particular we discuss the selection of variables and give a comparison with more traditional approaches such as discriminant analysis and the logit regression. The results demonstrate that the SVM has clear advantages over these methods for all variables tested.
    Keywords: Bankruptcy, Company rating, Default probability, Support vector machines
    JEL: C14 C45 G33
    Date: 2007
  5. By: Viktor Winschel; Markus Krätzig
    Abstract: We present a comprehensive framework for Bayesian estimation of structural nonlinear dynamic economic models on sparse grids. TheSmolyak operator underlying the sparse grids approach frees global approximation from the curse of dimensionality and we apply it to a Chebyshev approximation of the model solution. The operator also eliminates the curse from Gaussian quadrature and we use it for the integrals arising from rational expectations and in three new nonlinear state space filters. The filters substantially decrease the computational burden compared to the sequential importance resampling particle filter. The posterior of the structural parameters is estimated by a new Metropolis-Hastings algorithm with mixing parallel sequences. The parallel extension improves the global maximization property of the algorithm, simplifies the choice of the innovation variances, allows for unbiased convergence diagnostics and for a simple implementation of the estimation on parallel computers. Finally, we provide all algorithms in the open source software JBendge4 for the solution and estimation of a general class of models.
    Keywords: Dynamic Stochastic General Equilibrium (DSGE) Models, Bayesian Time Series Econometrics, Curse of Dimensionality
    JEL: C11 C13 C15 C32 C52 C63 C68 C87
    Date: 2008–02
  6. By: Andrew Feltenstein; Céline Rochon
    Abstract: In this paper, we study the impact of labor market restructuring and foreign direct investment on the banking sector, using a dynamic general equilibrium model with a financial sector. Numerical simulations are performed using stylized Chinese data, and banks failures are generated through increases in the growth rate of the labor force, a revaluation of the exchange rate or an increase in debt issue to finance the government deficit, as compared to a benchmark scenario in which banks remain solvent. Thus bank failures can result from what might seem to be either beneficial economic trends, or correct monetary and fiscal policies. We introduce fiscal policies that modify relative factor prices by lowering the capital tax rate and increasing the tax rate on labor. Such policies can prevent banking failures by raising the return to capital. It is shown that such fiscal policies are, in the short run, welfare reducing.
    Keywords: Banking failures; fiscal policies
    JEL: D58 E44 F37 G21
    Date: 2008
  7. By: Bill Gibson (University of Massachusetts Amherst)
    Abstract: Professor Crotty once casually observed that in his view economics could not be properly thought of as a science. This paper investigates the implications of this view in light of the question of how the scientific method has recently contributed to the evolution of economic practice. It is argued that agent-based models might provide a platform for an integration of recent micro and macroeconomic theories.
    Keywords: Agent-based models, macroeconomics, Keynes, James Crotty.
    Date: 2008–01
  8. By: Bill Gibson (University of Massachusetts Amherst)
    Abstract: Since the "closure debate" of the 1980s it is well known that com- parative static derivatives in analytical macro models are highly sensitive to the closure rule selected. This led Keynesians to conclude that Keynesian closures were superior to those favored by the orthodoxy and vice-versa. It is argued that with the advent of agent-based or multi-agent systems, the clo- sure debate is superseded. While elements of both Keynesian and neoclassical models survive the transition to the more synthetic environment, an agent- based approach eliminates the need for drastic simplification that was at the root of the debate from the beginning.
    Keywords: Agent-based models, multi-agent systems, macroeconomic closure.
    Date: 2008–01
  9. By: Dass Rajanish
    Abstract: Association rule mining is a well-known technique in data mining. Classification using association rules combines association rule mining and classification, and is therefore concerned with finding rules that accurately predict a single target (class) variable. The key strength of association rule mining is that all interesting rules are found. The number of associations present in even moderate sized databases can be, however, very large – usually too large to be applied directly for classification purposes. This project compares and combines different approaches for classification using association rules. This research area is called classification using association rules. An important aspect of classification using association rules is that it can provide quality measures for the output of the underlying mining process. The properties of the resulting classifier can be the base for comparisons between different association rule mining algorithms. A certain mining algorithm is preferable when the mined rule set forms a more accurate, compact and stable classifier in an efficient way. First, in this project we are interested in the comparison of the quality of different mining algorithms. Therefore, we use classification using association rules. Secondly, classification using association rules can be improved itself by using a mining algorithm that prefers highly accurate rules. The author of the report is indebted to several students and research assistants who showed interest and got involved in the work.
    Date: 2008–01–31
  10. By: Sang-Wook Stanley Cho (School of Economics, The University of New South Wales); Julian P. Diaz (Bowdoin College)
    Abstract: This paper analyzes the potential effects of two ongoing trade liberalization experiences: Ecuador signing a Free Trade Agreement with the United States and Slovenia joining the European Union as a full member. We construct a static Applied General Equilibrium Model and perform a numerical experiment that consists on eliminating all import tariffs that Ecuador and Slovenia impose on the United States and European Union, respectively. To calibrate our models, we work with Input-Output tables and construct a Social Accounting Matrix for each country. We perform additional numerical experiments, such as sensitivity analysis on the import and export elasticities of substitution, a partial liberalization scenario, the fiscal impact of eliminating the tariff revenues and how this loss can be compensated with other taxes, and an alternative trade liberalization framework for Slovenia. We find that both countries benefit from these trade liberalization reforms, with prices falling in the import sector and production rising in the export sector. However, different forms of trade liberalization (free trade agreement vs. customs union) have different implications on the patterns of trade and welfare.
    Keywords: Trade Liberalization; Free Trade Agreement; Customs Union; Fiscal Policy; Social Accounting Matrix; Ecuador; Slovenia
    JEL: F14 F15
    Date: 2007–08

General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.