Adaptive Polar Sampling (APS) is proposed as a Markov chain Monte Carlo method for Bayesian analysis of models with ill-behaved posterior distributions. In order to sample efficiency from such a distribution, location-scale transformation and a transformation to polar coordinates are used.

Le present rapport resume les perspectives nationales et provinciales etablies par Informetrica Limited pour le Systeme de projections des professions au Canada (SPPC) de Developpement des ressources humaines Canada. Les simulations relatives a l'economie nationale et aux economies provinciales...

Convention wisdom usually suggests that agents should use all the data they have to make the best possible prediction. In this paper it is shown that agents may sometimes be able to make better predictions by throwing away data. The optimality criterion agents adopt is the mean squared criterion.

This paper describes the application of a simulated annealing approach to the solution of a complex portfolio selection model. The model is a mixed integer quadratic programming problem which arises when Markowitz' classical mean-variance model is enriched with additional realistic constraints....

In this paper we test empirically with the Nordhaus and Yang (1996) RICE model the core property of the transfer scheme advocated by Germain, Toint and Tulkens (1997). This scheme is designed to sustain full cooperation in a voluntary international environmental agreement by making all countries...

L'objectif de cet article est de presenter quelques reflexions methodologiques relatives a la prevision econometrique des effets de regroupement de bureaux distributeurs dans le secteur postal. Deux concepts de fonction de cout sont utilises dans un but de comparaison: la fontion de cout espere...

We study the role of information sources on innovation decision in a two stage sequential probit model. In the first stage, firms decide to innovate or not, while in the second stage they choose the mode of innovation. We estimate parameters of this model in a Bayesian framework in which we use...

This paper presents information on the Stochastic Energy Deployment System (SEDS) model. DOE and NREL are developing this new model, intended to address many of the shortcomings of the current suite of energy models. Once fully built, the salient qualities of SEDS will include full probabilistic...