Catálogo de publicaciones - libros

The aim of this work is to present a new method of Value-at-Risk calculation using the fraction-of-time probability approach used in signal processing (see e.g Leśkow and Napolitano (2001)). This method allows making statistical type inferences based only on a single observation of phenomenon in time. Such setup is very convenient for time series data in financial analysis, when an assumption of having multiple realization of time series is very seldom realized. Another advantage of this method is the possibility of using it without assumptions on the distributions of returns. The aim of the paper is to present the method as well as application to financial data sets.

The last ten years have witnessed an increasing interest of the econometrics community in spectral theory. In fact, decomposing the series evolution in periodic contributions allows a more insightful view of its structure and of its cyclical behavior at different time scales. In this paper, the issues of cross-spectral analysis and filtering are concisely broached, dwelling in particular upon the windowed filter [15]. In order to show the usefulness of these tools, an application to real data — namely to US unemployment and inflation — is presented. By means of cross spectral analysis and filtering, a correlation can be found between these two quantities (i.e. the Phillips curve) in some specific frequency bands, even if it does not appear in raw data.

In this paper, we apply a dynamic microsimulation approach, which allows us to examine the evolution of the system as a whole and at the same time to focus our attention on the different typologies of workers and pensioners. The latter objective is achieved by simulating individual reactions to systemic changes, while taking into account the regional dimensions. This technique also enables us to perform a general micro-analysis of the effects of past reforms on family pension-income distribution and average individual pension-benefits. The analytical framework used is the dynamic microsimulation model MIND, jointly developed by the University of Parma and Pisa, that incorporates behavioural analysis of individual choices of retirement age, derived from the Stock-Wise [24] option value model. We also perform sensitivity analysis on the model. This represents a valuable technique for treating uncertainty in input variables and for testing the robustness of the simulation results to possible changes in the macroeconomic scenario. In particular, we measure the economic impact resulting from alternative values of the income growth rate and the real interest rate as well as the effect of different distributions of the education degree reached by the individuals.

The recent literature — including among others Redmond et al. [22], Gupta and Kapur [13], Mitton et al. [19] — has highlighted model alignment and validation as crucial issues to be tackled when microsimulating the consequences of public policies. This paper discusses some preliminary validation experiments performed on the model inputs, procedures and simulation results. The validation process that we use involves external checks, such as the ex-post comparison (from 1996 since 1999) of aggregated key macro variables (e.g. dependent workers’ incomes, and demographic variables) with official data (e.g. supplied by the Italian National Institute for Statistics ISTAT) at national and at regional level (North, Centre and South). We test the appropriateness of the main assumptions and specification of the model and policies’ effectiveness also using Monte Carlo simulations. Specifically, our analysis allows us to test MIND’s ability in forecasting demographic and economic trends and in capturing socio-economic dynamics for regional areas.

Recently, much attention has been devoted to econometric models as a new tool for dynamic microsimulation. In particular, microeconometric approaches to retirement decisions have been increasingly adopted for “calibrating” dynamic microsimulation frameworks aiming at endogenizing retirement choices. By doing this, both the understandment and the prediction of the effects of policy reforms (for istance, of Social Security systems) can be significantly improved. In this work an overview of the most recent developments in micromodeling retirement decisions is carried out. In particular, as for the choice of the estimation strategy, special emphasis is posed on the trade-off between the degree of realism of hypotheses, on the one hand, and on data tractability and/or estimation performance, on the other hand. Finally, some issues which represent a challenging avenue for future research are discussed.

The main objective of this paper is to illustrate, using Mexican data, how the results yield by modern econometric methods are dependent upon each specific technique as well as upon the statistical properties of the series analyzed. The problems are even stronger and more evident in the case of economic series with structural changes and high variability as is the case of Mexico. Applied econometrics should be explicitly based upon a probability viewpoint, and different methods should be taken to produce only approximations to the actual data generation process. Thus, alternative techniques can only show distinctive features of the actual data that still need to be validated with the rest of empirical evidence. This indicates that applied econometricians have to look for maximum information by correctly applying different techniques without forgetting the relevance of economic reasoning. Using Mexican data, alternative econometric estimations are evaluated indicating that the formulation of a monetary policy only on the basis of some specific technique, without considering its potential pitfalls, should not be recommended.

In this paper we extend the multi-regime framework to variables involved in the debate on economic growth and environmental quality, starting from a reexamination of the so-called Environmental Kuznets Curve. The aim is to discuss the double convergence hypothesis that implicitly stems from a recent line of research. According to it, some stylized facts would support the almost paradoxical hypothesis that economic growth produce not only cross-countries or regions convergence in per capita output, but also in (the demand of) environmental quality. Factual analysis seems to reject the hypothesis of convergence in output or income levels. Available evidence, rather, seems to point out that there is no such a thing as a unique avenue to sustainable development while the convergence predicted in more conventional analyses, in particular within the framework of the so called Environmental Kuznets Curve, is far away from being demonstrated. Actual growth processes do differ from each other in a deep qualitative sense, to the effect of profoundly influencing final outcomes as well as the unfolding of the processes themselves. This reflects differences in initial conditions, of course, but also the different sectoral or integrated policies that have been implemented along the way. Therefore, in contrast to the double convergence hypothesis , in our contribution we argue that growth is a necessary but not sufficient condition for the required change in the individuals preferences needed to shift social preferences away from private to public goods and that, moreover, the relationship between growth and environmental quality depends crucially upon the countrys growth model. Therefore, more than the quantitative it is the qualitative aspects that matters. The theoretical context that seems to lend itself to the analysis of such issues falls within the boundaries of the theories of endogenous growth. We argue that sustainable development, if it emerges at all, is the result of investment in immaterial capital (research, education and the like) more than the reflection of the exogenous forces (technological progress and demographic ) of the neoclassical theory. In the analysis of such issues, the environment offered by the multiregime approach proves useful as it highlights the qualitative properties of the dynamic processes, instead of focusing upon quantitative estimation of some special asymptotic states whose existence is often all but to be demonstrated.

This paper studies the determinants of growth rate volatility, focusing on the effect of level of GDP, structural change and the size of economy. First we provide a graphical analysis based on nonparametric techniques, then a quantitative analysis which follows the distribution dynamics approach. Growth volatility appears to (i) decrease with per capita GDP, (ii) increase with the share of the agricultural sector on GDP and, (iii) decrease with the size of the economy, measured by a combination of total GDP and trade openness. However, we show that the explanatory power of per capita GDP tends to vanish when we control for the size of the economy.

In the spirit of the New Tools perspective, this paper presents an overview that identifies the main changes in the measurement toolkit that have been used to test the empirical adequacy of the balance of payments constrained growth model (BPC-model) for comparative macroeconomic analysis as well as for the study of the long-term constraints on economic growth. It takes into account the BPC-model in its initial formulation, as put forward by A.P. Thirlwall nearly three decades ago, as well as in its recent developments introduced to account for alternative definitions of long-term equilibrium. It shows how the theoretical revisions of the BPC-model have been accompanied by a sophistication of its empirical testing techniques, better suited for the applied analysis of long-term economic relations. In addition, the toolkit currently used within this analytical perspective is illustrated with data for the Mexican economy, based on a version of the BPC-model that explicitly focuses on the relevance of the influence of interest payments on the economys long-term external equilibrium. The paper ends with some comments on the policy implications that may be derived from the application of the BPC-model, with special reference to the case of Latin America.

In the first part of this paper we present experimental evidence that the Mexican stock market has a fractal structure. We obtained the experimental results by using the Matsushita-Ouchi method, the box-counting method, and the fractal image compression technique of M. Barnsley. The results obtained by applying the Matsushita-Ouchi technique to the returns of the Indice de Precios y Cotizaciones (IPC) of the Mexican stock market support the assertion that the returns of the IPC behave like a fractional brownian motion. The box-counting method allows us to calculate the dimension of the graphs of the IPC, and its Hurst exponent H. The application of the fractal image compression technique produces attractors which are closed to the graphs of the returns of the IPC, and has permited us to estimate H. In the second part of the paper, we present two applications: first we study the efficiency graphs of the Mexican stock market, by plotting the Hurst exponent H as a funtion of time, and then we localize its structural changes by making use of a fractal attractor located in the phase space: H-IPC. We show that the fall of the IPC, that occurred between 1994 and 1995, corresponded to a rise of the value of H. We localize approximately the structural changes of the Mexican economy between 1987 and 1996.