giannone now casting care - university of notre dame
TRANSCRIPT
Panel on “Macroeconomic Forecasting and Nowcasting”
Domenico Giannone, Federal Reserve Bank of New York
2018 CARE Conference
Firm-level Information and the Macroeconomy
Disclaimer: The views expressed herein are solely those of the presenter and do not necessarily reflect those of the Federal Reserve Bank of New York or the Federal Reserve System.
Nowcasting
•Monitoring current economic conditions in real time
üModel-based counterpart to conjuctural analysis
üReal-time reading of the newsflow
üContinuously updated nowcast of GDP growth
Big-Data Analytics• High-dimensional data
üIncludes the large and complex data monitored by economists at central banks, trading desks, and in the media
• Entirely automatedüMimics best practice without relying on any judgment or subjective
prior information
• Real-timeüDigests new information within minutes of the releases
Digesting the Newsflow
•Coherent analysis of the link between macro news and cyclical developments
üExtract the news/surprise component from data� Actual data minus model-based forecasts
üTranslate the news in a common unit� What’s the impact of the news on GDP growth?
Methodology
•Dynamic factor modelü Few factors capture the salient features of business cycle
fluctuations� Flexibility, parsimony, robustness
• Filtering techniquesü Efficient processing of real-time information� Mixed frequencies, jagged edges, missing data
New York Fed Staff Nowcast
New York Fed Staff Nowcast
https://www.newyorkfed.org/research/policy/nowcast
New York Fed Staff Nowcast
New York Fed Staff Nowcast
Nowcasting 2016Q4
New York Fed Staff Nowcast
Nowcasting 2016Q4
New York Fed Staff Nowcast
Nowcasting 2016Q4
New York Fed Staff Nowcast
Nowcasting 2016Q4
New York Fed Staff Nowcast
Nowcasting 2016Q4
New York Fed Staff Nowcast
Nowcasting 2016Q4
New York Fed Staff Nowcast
Nowcasting 2016Q4
New York Fed Staff Nowcast
Nowcasting 2016Q4
üStatistical agencies already produce big and high-quality macroeconomic data
- Widely followed and intensively exploited
üIt takes work to bring new alternative data to comparable quality
- Distill signal from noise• Seasonality, outliers, other irregularities
- Reliability and replicability• Setting standards, quality control
• What types of data are useful in the context of nowcasting?• Is there a potential role for firm-level information?
Nowcasting
Jose M. Caribas (2018)“The real-time information content of macroeconomic news: implications for firm-level earnings expectations” Review of Accounting Studies
High-frequency Cash FlowsDavide Pettenuzzo,
Riccardo Sabbatucci,and Allan Timmermann
(2018)
High-frequency Cash FlowsDavide Pettenuzzo,
Riccardo Sabbatucci,and Allan Timmermann
(2018)
Further readingü Giannone et al. (2008) “Nowcasting: the Real-Time Information of
Macroeconomic Data Releases,” Journal of Monetary Economics
ü Stock and Watson (2017) “Twenty Years of Time Series Econometrics in Ten Pictures,” Journal of Economic Perspective
ü Bok et al. (2018) “Macroeconomic Nowcasting and Forecasting with Big Data,” NY Fed Staff Report (to appear on the Annual Review of Economics)
Caribas (2018) “The real-time information content of macroeconomic news: implications for firm-level earnings expectations,” Rev. Account. Stud.
Pettenuzzo et al. (2018) “High-frequency Cash Flows Dynamics,” Brandeis University mimeo