2018/08/09 by Leif Anders Thorsrud · 1 citation
Economics, Econometrics and Finance · #Complex Systems and Time Series Analysis #Financial Markets and Investment Strategies #Monetary Policy and Economic Impact
paper · doi:10.1080/07350015.2018.1506344
openalex publication_date 2018/08/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/29
I construct a daily business cycle index based on quarterly GDP growth and textual information contained in a daily business newspaper. The newspaper data are decomposed into time series representing news topics, while the business cycle index is estimated using the topics and a time-varying dynamic factor model where dynamic sparsity is enforced upon the factor loadings using a latent threshold mechanism. The resulting index classifies the phases of the business cycle with almost perfect accuracy and provides broad-based high-frequency information about the type of news that drive or reflect economic fluctuations. In out-of-sample nowcasting experiments, the model is competitive with forecast combination systems and expert judgment, and produces forecasts with predictive power for future revisions in GDP. Thus, news reduces noise. Supplementary materials for this article are available online.