Skip to content

Latest commit

 

History

History
7 lines (4 loc) · 1.72 KB

README.md

File metadata and controls

7 lines (4 loc) · 1.72 KB

This repository provides an analysis of the trend in forecast errors made by the Tealbook/Greenbook (GB) and the Survey of Professional Forecasters (SPF) for measures of the unemployment rate and real growth in personal consumption expenditures from 1982 to 2017. Below is a brief description of the main data and code used to produce the main results of this exercise.

The data on forecasts for unemployment and consumption made by the federal reserve (Tealbook/Greenbook) and the mean across private forcasters are provided by the Philidelphia Fed. Data on realized values of the forecasted variables are provided by the St. Louis Fed. The subfolder "data" contains the raw and cleaned versions of each of these datasets.

The subfolder "code/main" performs the "main" tasks of this exercise using the cleaned data. First, the GB and SPF data are used to compute a forecast for real growth in personal consumption expenditures starting in any given quarter. Next, the corresponding final measure of the same variable (PCE growth) is obtained from FRED. From there, the absolute value of the difference between the annual changes in observed and forecasted values is computed. Lastly, standard statistical techniques are used to analyze the trend in these GB and SPF absolute forecast errors.

The code file "reproduce.py" can be accessed to see the order in which these programs were run to produce the results. The code "reproduce.sh" will allow the user to reproduce the results from the terminal.