
on Computational Economics 
By:  Andreas S. Andreou (University of Cyprus); George A. Zombanakis (Bank of Greece) 
Abstract:  This paper applies computational intelligence methods to exchange rate forecasting. In particular, it employs neural network methodology in order to predict developments of the Euro exchange rate versus the U.S. Dollar and the Japanese Yen. Following a study of our series using traditional as well as specialized, nonparametric methods together with Monte Carlo simulations we employ selected Neural Networks (NNs) trained to forecast rate fluctuations. Despite the fact that the data series have been shown by the Rescaled Range Statistic (R/S) analysis to exhibit random behaviour, their internal dynamics have been successfully captured by certain NN topologies, thus yielding accurate predictions of the two exchangerate series. 
Keywords:  Exchange  rate forecasting, Neural networks 
JEL:  C53 
Date:  2006–11 
URL:  http://d.repec.org/n?u=RePEc:bog:wpaper:49&r=cmp 
By:  Mishra, SK 
Abstract:  In this paper we compare the performance of the Differential Evolution (DE) and the Repulsive Particle Swarm (RPS) methods of global optimization. To this end, seventy test functions have been chosen. Among these test functions, some are new while others are well known in the literature; some are unimodal, the others multimodal; some are small in dimension (no. of variables, x in f(x)), while the others are large in dimension; some are algebraic polynomial equations, while the other are transcendental, etc. FORTRAN programs of DE and RPS have been appended. Among 70 functions, a few have been run for small as well as large dimensions. In total, 73 optimization exercises have been done. DE has succeeded in 63 cases while RPS has succeeded in 55 cases. In almost all cases, DE has converged faster and given much more accurate results. The convergence of RPS is much slower even for lesser stringency on accuracy. Some test functions have been hard for both the methods. These are: ZeroSum (30D), Perm#1, Perm#2, Power and Bukin functions, Weierstrass, and Michalewicz functions. From what we find, one cannot reach at the definite conclusion that the DE performs better or worse than the RPS. None could assure a supremacy over the other. Each one faltered in some cases; each one succeeded in some others. However, DE is unquestionably faster, more accurate and more frequently successful than the RPS. It may be argued, nevertheless, that alternative choice of adjustable parameters could have yielded better results in either method’s case. The protagonists of either method could suggest that. Our purpose is not to join with the one or the other. We simply want to highlight that in certain cases they both succeed, in certain other case they both fail and each one has some selective preference over some particular type of surfaces. What is needed is to identify such structures and surfaces that suit a particular method most. It is needed that we find out some criteria to classify the problems that suit (or does not suit) a particular method. This classification will highlight the comparative advantages of using a particular method for dealing with a particular class of problems. 
Keywords:  : Global optimization; Stochastic search; Repulsive particle swarm; Differential Evolution; Clustering algorithm; Simulated annealing; Genetic algorithm; Tabu search; Ant Colony algorithm; Monte Carlo method; Box algorithm; NelderMead; Nonlinear programming; FORTRAN computer program; local optima; Benchmark; test functions 
JEL:  C61 C63 
Date:  2006–10–05 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1005&r=cmp 
By:  Yorgos Photis; Yorgos Grekousis 
Abstract:  The effectiveness of emergency service systems is measured in terms of their ability to deploy units and personnel in a timely, and efficient manner upon an eventâ€™s occurrence. A typical methodology to deal with such a task is through the application of an appropriate location  allocation model. In such a case, however, the spatial distribution of demand although stochastic in nature and layout, when aggregated to a specific spatial reference unit, appears to be spatially structured or semi â€“ structured. Aiming to exploit the above incentive, the spatial tracing and analysis of emergency incidents is achieved through the utilisation of Artificial Intelligence. More specifically, in the proposed approach, each location problem is dealt with at two interacting levels. Firstly, spatiotemporal point pattern of demand is analysed over time by a new genetic algorithm. The proposed genetic algorithm interrelates sequential events formulating moving objects and as a result, every demand point pattern is correlated both to previous and following events. Secondly, the approach provides the ability to predict, by means of an artificial neural network, how the pattern of demand will evolve and thus the location of supplying centres and/or vehicles can be optimally defined. The proposed neural network is also optimised through genetic algorithms. The approach is applied to Athens Metropolitan Area and the data come from Fire Departmentâ€™s records for the years 20032004. 
Date:  2006–08 
URL:  http://d.repec.org/n?u=RePEc:wiw:wiwrsa:ersa06p910&r=cmp 
By:  Silvio Giove (Department of Applied Mathematics, University of Venice); Paolo Bortot (Department of Applied Mathematics, University of Venice) 
Abstract:  In this paper we propose an iterative method to solve an optimal control problem, with fuzzy target and constraints. The algorithm is developed in such a way as to satisfy the target function and the constraints. The algorithm can be applied only if a method exists to solve a crisp parametric subproblem obtained by the original one. This is the case for a quadraticlinear target function with linear constraints, for which some well established solvable methods exist for the crisp associated subproblem. A numerical test confirmed the good convergence properties. 
Keywords:  fuzzy, mathematical programming 
JEL:  C6 
Date:  2006–11 
URL:  http://d.repec.org/n?u=RePEc:vnm:wpaper:148&r=cmp 
By:  Katerina Lisenkova; Peter Mcgregor; Nikos Pappas; Kim Swales; Karen Turner; Robert Wright 
Abstract:  This paper combines a multiperiod economic Computable General Equilibrium (CGE) modelling framework with a demographic model to analyse the macroeconomic impact of the projected demographic trends in Scotland. Demographic trends are defined by the existing fertilitymortality rates and the level of annual netmigration. We employ a combination of a demographic and a CGE simulation to track the impact of changes in demographic structure upon macroeconomic variables under different scenarios for annual migration. We find that positive net migration can cancel the expected negative impact upon the labour market of other demographic changes. (Pressure on wages, falling employment). However, the required size of the annual netmigration is far higher than the current trends. The policy implication suggested by the results is that active policies are needed to attract migrants. We nevertheless report results when varying fertility and mortality assumptions. The impact of varying those assumptions is rather small. 
Date:  2006–08 
URL:  http://d.repec.org/n?u=RePEc:wiw:wiwrsa:ersa06p432&r=cmp 
By:  Yuzuru Miyata; Hiroyuki Shibusawa; Yasuhiro Hirobata; Akira Ohgai 
Abstract:  In this paper, we aim to evaluate impacts of a new road network on the regional economy in Sanen, Japan. Sanen area is a boundary region between Aichi and Shizuoka Prefectures where the industrial sector is concentrated. The regional economy in Sanen strongly depends on the regional transportation networks. Recently, a new road construction is planned in the region. The efficiency of road investment is expected for the regional economy. We construct a spatial computable general equilibrium model to evaluate the border economy. The spatial economic impacts of a new road construction are analyzed by the numerical simulation under several scenarios. 
Date:  2006–08 
URL:  http://d.repec.org/n?u=RePEc:wiw:wiwrsa:ersa06p810&r=cmp 
By:  Lobianco, Antonello; Roberto, Esposti 
Abstract:  AgriPoliS is a multiagent mixed integer linear programming (MIP) model, spatially explicit, developed in C++ language and suitable for longterm sim ulations of agricultural policies. Beyond the mixed integer programming core, the model main feature is the interaction among a set of heterogeneous farm ers and between them and the environment in which they operate. In this paper we describe an extension of the model allowing AgriPoliS to deal with typical characters of the Mediterranean agriculture. In particular AgriPoliS was extended to allow a generic number of products and soil types, included perennial crops and products with quality differentiation. Furthermore, it can explicitly take into account irrigation. 
Keywords:  Mediterranean Agriculture; Common Agricultural Policy; Agentbased Models. 
JEL:  Q12 Q18 
Date:  2006–01 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1181&r=cmp 
By:  Lobianco, Antonello; Roberto, Esposti 
Abstract:  AgriPoliS is a multiagent mixed integer linear programming (MIP) model, spatially explicit, developed in C++ language and suitable for longterm simulations of agricultural policies. Once extended to deal with typical characters of the Mediterranean agriculture, AgriPoliS is used in this paper to describe the implementation of alternative policy cenarios and to apply them to two regions located in Central and South Italy. Results suggest that the effects of decoupling policies in the Mediterranean agriculture, as implemented in the 2003 reform, are often dominated by effects of structural trends and only a "bond scheme" would substantially change the regional farm structures. In no scenario we observe remarkable agricultural land abandonment. 
Keywords:  Mediterranean Agriculture; Common Agricultural Policy; MultiAgent Model 
JEL:  Q12 C61 Q18 
Date:  2006–09 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1182&r=cmp 
By:  Azzato, Jeffrey; Krawczyk, Jacek 
Abstract:  Computing the solution to a stochastic optimal control problem is difficult. A method of approximating a solution to a given stochastic optimal control problem using Markov chains was developed in [1]. This paper describes a suite of MATLAB functions implementing this method of approximating a solution to a given continuous stochastic optimal control problem. 
Keywords:  Computational techniques; Economic software; Computational methods in stochastic optimal control; Computational economics; Approximating Markov decision chains 
JEL:  C87 C63 
Date:  2006 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1179&r=cmp 
By:  Krawczyk, Jacek; Azzato, Jeffrey 
Abstract:  In this report, we outline a method for approximating a Markovian (or feedbackNash) equilibrium of a dynamic game, possibly subject to coupledconstraints. We treat such a game as a "multiple" optimal control problem. A method for approximating a solution to a given optimal control problem via backward induction on Markov chains was developed in Krawczyk (2006). A Markovian equilibrium may be obtained numerically by adapting this backward induction approach to a stage NikaidoIsoda function (described in Krawczyk & Zuccollo (2006)). 
Keywords:  Computational techniques; Noncooperative games; Econometric software; Taxation; Water; Climate; Dynamic programming; Dynamic games; Applications of game theory; Environmental economics; Computational economics; NikaidoIsoda function; Approximating Markov decision chains 
JEL:  C87 C63 Q25 C72 E62 
Date:  2006 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1195&r=cmp 
By:  Krawczyk, Jacek; Zuccollo, James 
Abstract:  A powerful method for computing Nash equilibria in constrained, multiplayer games is created when the relaxation algorithm and the NikaidoIsoda function are used together in a suite of MATLAB routines. This paper updates the MATLAB suite described in \cite{Berridge97} by adapting them to MATLAB 7. The suite is now capable of solving both static and openloop dynamic games. An example solving a coupled constraints game using the suite is provided. 
Keywords:  NikaidoIsoda function; Coupled constraints 
JEL:  C63 
Date:  2006–12 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1119&r=cmp 