Testing for Structural Change in Panel Data: Application to a Study of U.S. Foreign Trade in Manufacturing Goods

Abstract

A multivariate cusum test, along with extensions for partitioned parameters, serial correlation, and error components, is developed for panel data analysis. These techniques are then applied to a study of U.S. foreign trade in manufacturing goods examining changes in the detrminants, i.e., the relationships between the factor contents and the net exports.