Documentos de Trabajo publicados en 2000

• Lista de documentos de trabajo.
• Documentos de trabajo de otros años: 1993, 1994, 1995, 1996, 1997, 1998,1999 ,2001 ,2002 en adelante

Actividades del ICAE

Nota: Pulsando sobre el título se accede al resumen del trabajo. Pulsando sobre el autor se puede enviar un mensaje de correo electrónico.

"Vintage Capital ant the Dinamics of the AK Model". Raouf Boncekkine, Omer Licandro, Luis A. Puch y Fernando del Rio.

This paper analyzes the equilibrium dynamics of an AK-type endogenous growth model with vintage capital. The inclusion of vintage capital leads to oscilatiory dynamics governed by replacement echoes, which additionally influence the intercept of the balanced growth path. These features, which are in teh sharp contrast to those from standard AK model, can contribute to explaining teh short-run deviations observed between investment and growth rates time series. To characterize the convergence properties end the dynamics of the model we develop analytical and numerical methods thet should be of interest for the general resolution of endogenous growth models with vintage capital.

"Aplicaciones del Filtro de Kalman a las calibraciones en Modelos de Ciclo Real ". Jesús Ruiz.

Este trabajo tiene dos objetivos. El primero de ellos es aportar una generalización del filtro del Kalman a la estimación de modelos dinámicos con expectativas racionales formadas en el presente de variables endógenas futuras. El segundo es mostrar dos aplicaciones de este procedimiento en modelos estocásticos de crecimiento bajo el supuesto de expectativas racionales. En particular, se presenta, por un lado, una metodología para calibrar parámetros de estos modelos que resultan difíciles de estimar debido a la ausencia de datos en la economía real (por ejemplo, el coeficiente de aversión relativa al riesgo). El procedimiento que se presenta tiene la ventaja de la sencillez de su funcionamiento. Por otro lado, se utiliza el procedimiento de calibración para dar una medida objetiva de discriminación entre modelos, que permita resolver el problema de identificación de modelos observacionalmente equivalentes.

"Efficient Solution Concepts and Their Relations in Stochastic Multiobjective Programming". R. Caballero, E. Cerdá, M. M. Muñoz, L. Rey e I. Stancu-Minasian.

In this work different concepts of efficient solutions to problems of Stochastic multiple objective programming are analyzed. We centre our interest on poblems in which some of the objective functions We propose a test for multivariate seasonality. Starting from a multivariate model for the seasons, some constraints moust hold both, on the covariance matrix of the innovations, as well as among coefficients across equations for a univariate representati of seasonality to be apropriate. Applied to a set of U.K. macroeconomic variable, our test shows that a multivariate representation of seasonality should generally be preferred.

"An Optimal Sequence of Landfills in Municipal solid Waste Management". Francisco Javier André y Emilio Cerdá.

Given that landfills are depletable and replaceable resources, the right approach, when dealing with landfill management, is that of designing an optimal sequence of landfills rather than every single landfill separately. In this paper we use Optimal Control models, with mixed elementsof both continuous and discrete time problems, to determine an optimal sequence of landfills, as regarding their capacity and lifetime. The resulting optimization problems involve dividing a time horizont of planning into subintervals the length of which has to be dicided. In each of the subintervals some cost, the ammount of which depends on the value of the decision variable, have to borne. The obtained results are useful for other economic problems such as private and public investment, consumption decisions on durable goods, etc.  

"A MATLAB Toolbox for Reliable Time Series Modeling and Forecasting in State-Space". Jaime Terceiro, José Manuel Casals, Miguel Jerez y Sonia Sotoca.

Software reliability is a wide issue, depending not only on the use of stable implementations of well-reputed algorithms, but also on software desing aspects. This philosophy is implemented in E⁴, a MATLAB Toolbox which uses state-space methods to achive both, flexibility and reliability. E⁴ support many standard formulations such as VARMAX, econometric models in structural form, transfer functions or general linear state-space models. These models are estimated by exact maximun-likelihood, under standard conditions, or in an extended framework that allows for measurement errors, missing data, vector GARCH errors and constraints on the parameters. Ready-to-use functions are provided for model especification, preliminary estimation by subespace methods, analytic computation of the likelihood gradient and information matrix, simulation, forecasting and signal extraction. The core algorithms have been optimized for stability and numerical accuracy. In this aspect, the use of a computational platform such as MATLAB guarantees computational accuracy, portability and consistency between hardware-software platforms. Several examples illustrate the main features of the Toolbox.

"An Exact Multivariate Model-Based Structural Decomposition". Jose Manuel Casals, Miguel Jerez y Sonia Sotoca.

We describe a simple procedure for decomposing a vector of time series into trend, cycle, seasonal and irregular components. Contrary to common practice, we do not assume these components to be orthogonal conditional to their past. However, the state-space representation employed assures that their smoothed estimates converge to exact values, with null variances and covariances. Among other implications, this means that the component are not revised when the samplle increases. The practical applications of the method is ilustrated both with simulated and real data.

"The Evolution of the ph in Europe 1986-1997, with Panel Data" E. Martín, T. Pérez-Amaral, A. Rua y E. Hernández.

The temporal evolution of pH values in precipitation over Europe during the period 1986-1997 is examined using panel data. The use of panel data techniques allows us to detrmine the themporal evolution of groups of stations rather than analysing the temporal behavior of each of them. The analysis reveals three different temporal patterns: Peripheral, Central and French. We find a significant increasing trend (p < 0.00001) in both Peripheral and Central Patterns. The annual increases are +0.057 pH-units yr^-1 and +0.022 pH-units yr^-1 respectively. However the French Pattern is characterized by a significant decreasing trend (p < 0.004) and the annual decreases is -0.022 pH-units yr^-1. The standard errors of panel data estimates are around 47% smaller then those of classical pooling and 32% smaller than aggregate time series regression.The use of panel data produces higher R² values than classical pooling and aggregate daa. This technique takes into account the individual heterogeneity, allows a larger number of data points and improves the efficiency of the estimates. On general, the policies of governments to reduce pollutant emissions siems to be effective.

Actividades del ICAE

Instituto Complutense de Análisis Económico