Working Paper: CEPR ID: DP5829
Authors: Christine De Mol; Domenico Giannone; Lucrezia Reichlin
Abstract: This paper considers Bayesian regression with normal and double exponential priors as forecasting methods based on large panels of time series. We show that, empirically, these forecasts are highly correlated with principal component forecasts and that they perform equally well for a wide range of prior choices. Moreover, we study the asymptotic properties of the Bayesian regression under Gaussian prior under the assumption that data are quasi collinear to establish a criterion for setting parameters in a large cross-section.
Keywords: Bayesian VAR; Large Cross-sections; Lasso Regression; Principal Components; Ridge Regressions
JEL Codes: C11; C13; C33; C53
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Bayesian regression forecasts under normal and double-exponential priors (C11) | PCR forecasts (C59) |
| Bayesian methods (C11) | forecast accuracy (C53) |
| Cross-section size (n) and sample size (t) approaching infinity (C13) | Bayesian regression forecasts converge to efficient forecast (C53) |
| Double-exponential prior (C46) | recovery of large coefficients (C55) |
| Bayesian regression forecasts (C53) | forecasting effectiveness (C53) |