Journal Article

The Chen-Tindall system and the lasso operator: improving automatic model performance


Abstract: Using U.S. monthly macroeconomic data, the automatic model system presented in Chen and Tindall [2016] outperforms the lasso automatic system, but the lasso is improved where Bayesian model averaging is employed to combine its forecasts with those from autoregressive schemes. The best performance is obtained using Bayesian model averaging to combine the Chen?Tindall system, the lasso, and autoregressive schemes. Performance is virtually the same using this combined approach where the elastic-net operator is substituted for the lasso. Similar overall outcomes are found for France and Germany treated as a single economic system and for Canada.

Keywords: Automatic model building; Bayesian model averaging; the elastic net.; the lasso;

Access Documents

File(s): File format is application/pdf http://www.dallasfed.org/assets/documents/banking/occasional/1601.pdf
Description: Full text

Authors

    Chen, Jiaqi

    Tindall, Michael

Bibliographic Information

Provider: Federal Reserve Bank of Dallas

Part of Series: Occasional Papers

Publication Date: 2016-05-31

Pages: 19 pages

Order Number: 16-1