Similar Search Results

Instabilité de la courbe de Phillips aux Etats-Unis : un modèle explicatif à changements de régimes.

Guerrero, Guillaume; Million, Nicolas - Maison des Sciences Économiques, Université Paris 1 … - 2004

This paper proposes a new empirical representation of inflation expectations errors in a Space-State Markov-Switching framework. We explicitly identify the dynamics of inflation expectation errors using the expectations augmented Markov-Switching Phillips curve as a measurement equation. In this...

Persistent link: https://www.econbiz.de/10005797809

Finite-Sample Simulation-Based Inference in VAR Models with Applications to Order Selection and Causality Testing

Dufour, Jean-Marie; Jouini, Tarek - Centre Interuniversitaire de Recherche en Analyse des … - 2005

Statistical tests in vector autoregressive (VAR) models are typically based on large-sample approximations, involving the use of asymptotic distributions or bootstrap techniques. After documenting that such methods can be very misleading even with fairly large samples, especially when the number...

Persistent link: https://www.econbiz.de/10005100698

Distribution-Free Bounds for Serial Correlation Coefficients in Heteroskedastic Symmetric Time Series

Dufour, Jean-Marie; Farhat, Abdeljelil; Hallin, Marc - Centre Interuniversitaire de Recherche en Analyse des … - 2005

We consider the problem of testing whether the observations X1, · · ·, Xn of a time series are independent with unspecified (possibly nonidentical) distributions symmetric about a common known median. Various bounds on the distributions of serial correlation coefficients are proposed:...

Persistent link: https://www.econbiz.de/10005100838

Short Run and Long Run Causality in Time Series: Inference

Dufour, Jean-Marie; Pelletier, Denis; Renault, Éric - Centre Interuniversitaire de Recherche en Analyse des … - 2003

techniques as well as standard Gaussian asymptotic distributional theory. Bootstrap procedures are also considered. For the case … régression linéaire de même que la théorie distributionnelle asymptotique gaussienne habituelle. Dans le cas des processus … intégrés, nous proposons des méthodes de régression étendue qui ne requièrent pas de théorie asymptotique non standard. L …

Persistent link: https://www.econbiz.de/10005100843

Markovian Processes, Two-Sided Autoregressions and Finite-Sample Inference for Stationary and Nonstationary Autoregressive Processes

Dufour, Jean-Marie; Torrès, Olivier - Centre Interuniversitaire de Recherche en Analyse des … - 2000

In this paper, we develop finite-sample inference procedures for stationary and nonstationary autoregressive (AR) models. The method is based on special properties of Markov processes and a split-sample technique. The results on Markovian processes (intercalary independence and truncation) only...

Persistent link: https://www.econbiz.de/10005100872