Model uncertainty in cross-country growth regressions

By C. Fernández, E. Ley, Mark F. J. Steel. 2001

897 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/83fc55554908b5f45d4e075daec2d6dbc31d1ff9

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

Citations in the corpus

Tourism and Economic Growth: A Meta-regression Analysis (2019, background): “This is conceptually known as model uncertainty (Fernandez, Ley, and Steel 2001).”
Forecasting U.S. Inflation by Bayesian Model Averaging (2003, neutral): “of the prior for the error variance, that was made by Fernandez, Ley and Steel (2001) and”
Bayesian Model Averaging and exchange rate forecasts (2003, neutral): “Sala-i-Martin (2000) and Fernandez, Ley and Steel (2001) ) and stock return prediction”
Least Squares Model Averaging (2007, neutral): “Taking conditional expectations of (18) we obtain”
Trust and growth: a shaky relationship (2008, neutral): “The fixed variables are control variables that are 13 Recently, robustness tests based on Bayesian analysis have been proposed and used – see e.g. Fernandez et al. (2001) and Sala-i-Martin et al. (2004).”
Complete subset regressions (2013, neutral): “We follow Fernandez et al. (2001a) and set g to 1/T or 1/K 2.”
Bayesian Model Averaging and Weighted-Average Least Squares: Equivariance, Stability, and Numerical Issues (2011, positive): “This leads to a conditional joint prior distribution of the form where V 0 i is the variance-covariance matrix of the prior distribution of β 2 i which takes the standard form proposed by Zellner (1986) and Fern´andez et al. (2001) and g = 1 / max( n, k 22 ) is a constant scalar for each model M i .”
Time-varying sparsity in dynamic regression models (2014, positive): “We use the default choices of Fernandez et al. (2001) for the g-prior with the previous k observations, i.e. yt−k, . . . , yt−1, to predict at each time point.”
Forecast Combination and Bayesian Model Averaging - A Prior Sensitivity Analysis (2012, methodology): “For the simulation study we design two different settings: Setup ’A’ follows closely Fernández et al. (2001a) in drawing first 10 potential explanatory variables (x1, . . . , x10) using N =120 draws from a standard normal distribution for the covariates.”