Comments on ‘Forecasting with a real-time data set for macroeconomists’ |
| |
Authors: | Sharon Kozicki |
| |
Institution: | Research Division, Federal Reserve Bank of Kansas City, 925 Grand Boulevard, Kansas City, MO 64198, USA |
| |
Abstract: | The effects of data revision as removing measurement error on forecast model building is examined. I show various effects on model building using revised or instead real-time data. These effects include lag length selection and measured persistence. I then argue that in practice that one should include the entire dataset available (real-time and past revised data) in forecast construction. |
| |
Keywords: | Forecasting Real-time data Measurement error Data revision |
本文献已被 ScienceDirect 等数据库收录! |