Premium
Improving Local Manufacturing Employment Forecasts Using Cointegration Analysis
Author(s) -
Crane Steven E.,
Nourzad Farrokh
Publication year - 1998
Publication title -
growth and change
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.657
H-Index - 55
eISSN - 1468-2257
pISSN - 0017-4815
DOI - 10.1111/1468-2257.00082
Subject(s) - cointegration , economics , process (computing) , manufacturing sector , error correction model , econometrics , macroeconomics , adaptation (eye) , business cycle , computer science , physics , optics , operating system
Procedures for tracking and forecasting economic conditions in regional economies have evolved significantly over the last 30 years. Much of this evolution has followed developments in macroeconomics, where techniques for tracking/forecasting key economic variables have tended to originate. This technique adoption and adaptation process continues today, as developments in the technique adoption and adaptation process continues today, as developments in the modeling of cointegrated macroeconomic time series have begun to appear in the regional modeling and forecasting literature. This paper presents an effort at modeling a segment of a regional economy using the cointegration testing procedures suggested by Johansen and Jusilius (1990) to develop a forecasting model for manufacturing employment in Milwaukee, WI. The paper demonstrates how Vector Error Correction (VEC) modeling can lead to gains in the accuracy of local manufacturing employment forecasts relative to more traditional VAR models in either levels or first‐differenced form. In the process, it demonstrates procedures for developing a relatively simple VEC model that reveals something about the structure of the local manufacturing sector, including possible linkages to the national economy. This information can assist local policy makers in anticipating and adapting to business cycle‐related fluctuations in this critical sector of the local economy.