Le propos de cet article est une evaluation critique des methodes de rapprochement entre les faits et la theorie economique. Il s'agit, plus precisement, de questionner les methodes econometriques actuellement en vigueur relativement a l'epistemologie scientifique. Nous avons considere dans ce...

We propose a test for spatial correlation in Probit models that is a joint test for exclusion of spatially lagged-dependent variables and absence of spatial-error correlation. We give a maximum-likelihood justification for the test but use a simulations approach rather than relying on its...

We propose a set of algorithms for testing the ergodicity of empirical time series, without reliance on a specific parametric framework. It is shown that the resulting test asymptotically obtains the correct size for stationary and nonstationary processes, and maximal power against non-ergodic...

We provide numerically reliable analytical expressions for the score of conditionally heteroskedastic dynamic regression models when the conditional distribution is multivariate t. We also derive one-sided and two-sided LM tests for multivariate normality versus multivariate t based on the first...

In this paper we consider the problem of testing the null hypothesis that a series has a constant level against the alternative that the level follows a random walk. This problem has previously been studied by inter alia, Nyblom and Makelainen in the context of the orthogonal random walk plus...

This paper compares and generalizes some testing procedures for structural change in the context of cointegrated regression models.

We consider data on jewellery sold in English public auctions between June 1993 and May 1994 at Credit Municipal de Paris. We present the underlying model of this market derived from a "hedonic price equation".

In this paper we propose a new test procedure with more general steady state information to test the convergence hypothesis for a specific economy. We consider a model where demeaned per capita output of an economy is a function of time trend and then set the convergence hypothesis as negative...

This note gives a few practical guidelines for cointegration analysis. The focus is on testing the cointegration rank in a VAR model and on how an intercept and a trend should be incorporated in the model. Only two cases appear relevant for most economic data.