Time-Varying Dictionary and the Predictive Power of FED Minutes

47 Pages Posted: 18 Jan 2019

See all articles by Luiz Renato Lima

Luiz Renato Lima

University of Tennessee, Knoxville

Lucas Godeiro

Federal Rural University of Semi-Arid - UFERSA

Mohammed Mohsin

University of Tennessee, Knoxville - College of Business Administration - Department of Economics

Date Written: January 9, 2019

Abstract

This paper develops a novel method to extract the most predictive information from FED minutes. Instead of considering a dictionary (set of words) with a fixed content, we construct a dictionary whose content is allowed to change over time. Specifically, we utilize machine learning to identify the most predictive words (the most predictive content) of a given minute and use them to derive new predictors. We show that the new predictors improve real time forecasts of output growth by a statistically significant margin, suggesting that the combination of supervised machine learning and text regression can be interpreted as a powerful device for out-of-sample macroeconomic forecasting.

Keywords: Text Regression, Supervised Machine Learning, Elastic Net, Central Bank Communication, Forecasting, Real Time

JEL Classification: 53, C55, E37, E47

Suggested Citation

Lima, Luiz Renato and Godeiro, Lucas and Mohsin, Mohammed, Time-Varying Dictionary and the Predictive Power of FED Minutes (January 9, 2019). Available at SSRN: https://ssrn.com/abstract=3312483 or http://dx.doi.org/10.2139/ssrn.3312483

Luiz Renato Lima (Contact Author)

University of Tennessee, Knoxville ( email )

Knoxville, TN 37996
United States

Lucas Godeiro

Federal Rural University of Semi-Arid - UFERSA ( email )

MossorĂ³
Brazil
+55-3317-8555 (Phone)

HOME PAGE: http://www.ufersa.edu.br

Mohammed Mohsin

University of Tennessee, Knoxville - College of Business Administration - Department of Economics ( email )

508 Stokely Management Center
Knoxville, TN 37996-0550
United States
865-974-1690 (Phone)
865-974-4601 (Fax)

HOME PAGE: http://econ.bus.utk.edu/Mohsin/mohsin.htm

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
123
Abstract Views
776
rank
254,669
PlumX Metrics