Discovering structural break in precious metal time series data

摘要

Structural break is an important issue in macroeconomic time series data. The aim of this paper is to examine the structural break and to determine the exact break date in the price of commodity data using monthly data. We used two different techniques; first is Unit Root with Structural Break procedure and secondly is Zivot and Andrews test that allows for detecting a break at an unknown date. The results manage to detect structural break with different break date for different technique used during the period 1989 to 2009. We also conduct Bai Perron test which allows for more than one structural break. The result show that we manage to detect the best two break date for all three commodities.