Premium
Measuring the Information Content of the Beige Book: A Mixed Data Sampling Approach
Author(s) -
ARMESTO MICHELLE T.,
HERNÁNDEZMURILLO RUBÉN,
OWYANG MICHAEL T.,
PIGER JEREMY
Publication year - 2009
Publication title -
journal of money, credit and banking
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.763
H-Index - 108
eISSN - 1538-4616
pISSN - 0022-2879
DOI - 10.1111/j.1538-4616.2008.00186.x
Subject(s) - predictive power , content (measure theory) , aggregate (composite) , schedule , sampling (signal processing) , econometrics , computer science , economics , management , mathematics , telecommunications , philosophy , mathematical analysis , materials science , epistemology , detector , composite material
Studies of the predictive ability of the Federal Reserve's Beige Book for aggregate output and employment have proven inconclusive. This might be attributed, in part, to its irregular release schedule. We use a model that allows for data sampling at mixed frequencies to analyze the predictive power of the Beige Book. We find that the Beige Book's national summary and District reports predict GDP and aggregate employment and that most District reports provide information content for regional employment. In addition, there appears to be an asymmetry in the predictive content of the Beige Book language.