nep-cmp New Economics Papers
on Computational Economics
Issue of 2008‒01‒12
six papers chosen by
Stan Miles
Thompson Rivers University

  1. The Macroeconomic, Industrial and Distributional Effects of Removing Tariffs in Bangladesh By Serajul Hoque
  2. Parallelization of Matlab codes under Windows platform for Bayesian estimation: A Dynare application By Ivano Azzini; Riccardo Girardi; Marco Ratto
  3. Independent Component Analysis Via Copula Techniques By Ray-Bing Chen; Meihui Guo; Wolfgang Härdle; Shih-Feng Huang
  4. Do repeated game players detect patterns in opponents? Revisiting the Nyarko & Schotter belief elicitation experiment By Spiliopoulos, Leonidas
  5. Using Statistics Canada LifePaths Microsimulation Model to Project the Health Status of Canadian Elderly By Jacques Légaré; Yann Décarie
  6. The Bayesian Additive Classification Tree Applied to Credit Risk Modelling By Junni L. Zhang; Wolfgang Härdle

  1. By: Serajul Hoque
    Abstract: This paper examines the economic effects of removing tariffs in Bangladesh using a computable general equilibrium (CGE) modelling approach. The results of the simulations indicate that in the short-run a funded tariff cut with fixed real national savings would increase employment slightly and hence would expand GDP. There would be a small economy-wide welfare gain as measured by real consumption. The sectoral results showed that export-oriented industries would experience an expansion in output and employment. There also would be positive effects on the suppliers to these industries. Lightly-protected industries, which rely heavily on imported intermediate inputs, are projected to show robust expansion as they would benefit from a cost reduction. However, highly-protected, import-competing industries would suffer a contraction in output and employment as they would face increased competition from imports due to the removal of tariffs. The simulation results also indicate that there would have some noticeable effects on the distribution of real consumption between different household groups. Overall, urban households would experience an expansion in real consumption and rural households would suffer a contraction as a consequence of the funded tariff cut with fixed real national savings.
    Keywords: CGE model, trade liberalisation, income distribution, Bangladesh
    JEL: C68 F13 O15
    Date: 2008–01
  2. By: Ivano Azzini; Riccardo Girardi; Marco Ratto (Euro-area Economy Modelling Centre)
    Abstract: In the Bayesian estimation of DSGE models with DYNARE (specifically the Matlab Version for Windows), most of the computing time is devoted to the posterior estimation with the Metropolis algorithm. Usually, the Metropolis is run using multiple parallel chains, to allow more careful convergence testing. In this work we describe a way to parallelize the multiple-chain Metropolis algorithm within the Dynare Framework, running parallel chains on different processors to reduce computational time. To do so, we aimed at the easiest and cheapest possible strategy, i.e. the one which requires the lower level of modification in the basic DYNARE routines and does not need any licensed toolbox. Despite the fact that parallelizing the Metropolis-Hasting algorithm is intrinsically easy (the different chains are completely independent each other and do not require communication between them), MatLab software does not allow concurrent programming, or in other terms it does not support multi-threads, without the use of MatLab Distributed Computing Toolbox. The general idea is that when the execution of the Metropolis should start, instead of running it inside the MatLab session, the control of the execution is passed to the operating system that allows for multi-threading and concurrent threads are launched on different processors. When the metropolis computations are concluded the control is given back to the original MatLab session for post-processing Markov Chain results.
    Keywords: North-South, DSGE models, DYNARE, Matlab, Windows, Parallel Computing
    JEL: C63
    Date: 2007
  3. By: Ray-Bing Chen; Meihui Guo; Wolfgang Härdle; Shih-Feng Huang
    Abstract: Independent component analysis (ICA) is a modern factor analysis tool de- veloped in the last two decades. Given p-dimensional data, we search for that linear combination of data which creates (almost) independent components. Here copulae are used to model the p-dimensional data and then independent components are found by optimizing the copula parameters. Based on this idea, we propose the COPICA method for searching independent components. We illustrate this method using several blind source separation examples, which are mathematically equivalent to ICA problems. Finally performances of our method and FastICA are compared to explore the advantages of this method.
    Keywords: Blind source separation, Canonical maximum likelihood method, Givens rotation matrix, Signal/noise ratio, Simulated annealing algorithm
    JEL: C01 C13 C14 C63
    Date: 2008–01
  4. By: Spiliopoulos, Leonidas
    Abstract: The purpose of this paper is to reexamine the seminal belief elicitation experiment by Nyarko and Schotter (2002) under the prism of pattern recognition. Instead of modeling elicited beliefs by a standard weighted fictitious play model this paper proposes a generalized variant of fictitious play that is able to detect two period patterns in opponents’ behavior. Evidence is presented that these generalized pattern detection models provide a better fit than standard weighted fictitious play. Individual heterogeneity was discovered as ten players were classified as employing a two period pattern detection fictitious play model, compared to eleven players who followed a non-pattern detecting fictitious play model. The average estimates of the memory parameter for these classes were 0.678 and 0.456 respectively, with five individual cases where the memory parameter was equal to zero. This is in sharp contrast to the estimates obtained from standard weighted fictitious play models which are centred on one, a bias introduced by the absence of a constant in these models. Non-pattern detecting fictitious play models with memory parameters of zero are equivalent to the win-stay/lose-shift heuristic, and therefore some sub jects seem to be employing a simple heuristic alternative to more complex learning models. Simulations of these various belief formation models show that that this simple heuristic is quite effective against other more complex fictitious play models.
    Keywords: learning; game theory; behavioral game theory; fictitious play; repeated games; mixed strategy; non-cooperative games; pattern recognition; pattern detection; experimental economics; beliefs; belief elicitation; strategic
    JEL: C9 C63 C73 C72
    Date: 2008–01–09
  5. By: Jacques Légaré; Yann Décarie
    Abstract: Complex population projections usually use microsimulation models; in Canada, Statistics Canada has developed a global dynamic microsimulation model named LifePaths in the Modgen programming language to be used in policy research. LifePaths provides a platform to build on for our research program, conjointly with Dr Janice Keefe from Mount Saint Vincent University, on projections of the Canadian chronic homecare needs for the elderly up to 2031 and of the human resources required. Beside marital status, family networks and living arrangements, future health status of the elderly is a key variable, but an intricate one. Since health status transitions were previously conditioned only on age and sex, we will use here the current disability module of LifePaths with longitudinal data from Canada’s National Population Health Survey (NPHS). These new health status transitions are considering other significant explicative variables like marital status, education etc. We will then present projections of future Canadian elderly by health status and a comparison with nine European countries for the Future Elderly Living Conditions in Europe (FELICIE) Research Program which has used the same approach. Our previous researches have shown the importance of future disability level for the management of an elderly society. The main output of the present paper would first produce, with new health scenarios, new estimates for Canada of elderly in poor health, for those aged 75 and over. Secondly, it would produce an interesting comparative analysis, useful especially for implementing new policies for the well-being of the Canadian elderly.
    Keywords: Microsimulation, Elderly population, Aging, LifePaths, Health, Canada
    JEL: C15 I19
    Date: 2008–01
  6. By: Junni L. Zhang; Wolfgang Härdle
    Abstract: We propose a new nonlinear classification method based on a Bayesian "sum-of-trees" model, the Bayesian Additive Classification Tree (BACT), which extends the Bayesian Additive Regression Tree (BART) method into the classi- fication context. Like BART, the BACT is a Bayesian nonparametric additive model specified by a prior and a likelihood in which the additive components are trees, and it is fitted by an iterative MCMC algorithm. Each of the trees learns a different part of the underlying function relating the dependent variable to the input variables, but the sum of the trees offers a flexible and robust model. Through several benchmark examples, we show that the BACT has excellent performance. We apply the BACT technique to classify whether firms would be insolvent. This practical example is very important for banks to construct their risk profile and operate successfully. We use the German Creditreform database and classify the solvency status of German firms based on financial statement information. We show that the BACT outperforms the logit model, CART and the Support Vector Machine in identifying insolvent Firms.
    Keywords: Classi¯cation and Regression Tree, Financial Ratio, Misclassification Rate, Accuracy Ratio
    JEL: C14 C11 C45 C01
    Date: 2008–01

This nep-cmp issue is ©2008 by Stan Miles. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
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.