Jean-Louis Brillet

Measuring forecast uncertainty: A review with evaluation based on a macro model of the French economy

Abstract Five alternative techniques have been applied to measure the degree of uncertainty associated with the forecasts produced by a macro-model of the French economy, the Mini-DMS developed at INSEE. They are bootstrap, analytic simulation on coefficients, Monte Carlo on coefficients, parametric stochastic simulation and re-estimation, a residual-based procedure. Due to the complexity and the size of the model (nonlinear and with more than 200 equations), several associated technical problems had to be solved. The remarkable convergence of results which has been obtained for all the main endogenous variables suggests that forecast confidence intervals are likely to be quite reliable for this model.

Analyse et mesure de l'incertitude en prevision d'un modele econometrique. Application au modele mini-DMS

This article describes the application to an operational medium-size econometric model, mini-DMS, of methods associating, to deterministic forecasts, a measure of the uncertainty due to the stochastic nature of behavioural equations. After having described the theoretical and practical foundations of the methods, we shal l analyze sequentially the deterministic bias, the uncertainty (standard error) of forecasts and of policy instruments, trying to look at the information from the point of view of the policy maker.This article describes the application to an operational medium-size econometric model, mini-DMS, of methods associating, to deterministic forecasts, a measure of the uncertainty due to the stochastic nature of behavioural equations. After having described the theoretical and practical foundations of the methods, we shal l analyze sequentially the deterministic bias, the uncertainty (standard error) of forecasts and of policy instruments, trying to look at the information from the point of view of the policy maker.

Solving Large and Small Models on Microcomputers

In this paper we evaluate available packages designed for macroeconomic modelling. First we stress the qualities needed for building, maintaining and using an operational model, used professionally for forecasting purposes. We shall first address the global features of the process, then its different steps: gathering data, establishing equations, estimating the behavioral ones, solving the system, and presenting the results. Then, we assess the individual qualities of available packages and their suitability to meet these tasks. As simulation is probably the most often repeated task in the modelling process, we pay particular attention to the speed and versatility of the algorithm used, to its probability of reaching a solution, and to the features offered in case of trouble. The packages in the survey are AREMOS for Windows, published by WEFA, SA (version 5.1) Portable TROLL, distributed by INTEX (USA) and HendyPlan (version 1.79) SORITEC for Windows 95/NT, produced by the Sorites Group, Inc. (version 1.01). We base our tests on the 501-equation AMADEUS model, used officially by INSEE to produce its medium term forecasts. Tables give statistics for each basic modelling task: database handling, model specification, estimation, and simulation. Primarily we consider speed, but also the space taken by the associated files. Further, we evaluate the efficiency of our three options, combining the above results with more subjective observations of features provided. In our opinion, no package dominates the others. It all depends on the nature of the modelling task. Portable Troll should be the best option for the practical scientist, who wants to simulate complex models and apply complex methods. User-friendliness is not the main priority, and the display of results can be trusted to other packages. AREMOS will satisfy business economists, who probably do not need sophisticated scientific options but require quick results through user-friendly features, which should compare with the other packages they use. SORITEC is less powerful than TROLL and much less user-friendly than AREMOS, but it is also much less expensive. It presents clearly the best quality-price ratio, especially for solving medium-size problems.

Uncertainty and Stability in a Macro-Econometric Model

One of the main techniques for determining the long term stability properties of a macro-econometric model has been for some time now to compute the eigenvalues of the linearized reduced form of the model. But these eigenvalues are affected by uncertainty, coming mostly from the error on the estimated coefficients. In this paper we study, using the French macro-economic model Mini-DMS, how taking into account the uncertainty can affect the conclusions. We shall give particular attention to the following points: are the conclusions concerning convergence certain, do the eigenvalues change significantly with the period at which the linearization is made, does the use of elasticities instead of multipliers produce significantly more stable eigenvalues.