Authors: Dimitris Pachis

Title: A structural break or a shock with extended half-life?

Abstract

A major issue in the literature of time series is the identification of the nature of a shock. The emergence of the autoregressive models and their co-integrated counterparts aimed at identifying whether a shock is of permanent or transitory nature. This identification was based on testing the stationarity of the series, followed by the meta-analysis of Granger causality, impulse responses and forecast error decomposition. However, the limited length of a time series always sets the question whether the identified response of the economic actors was due to a permanent shock as implied by a unit root or a shock with extended half-life. In this case, the panel autoregressive models can shed more light to the data generating process. This paper attempts to investigate this issue by modeling the price transmission mechanism of fresh tomatoes in Greece between producer and consumer with both approaches. The aim is to discover whether the traditional estimation method of vector autoregressions is misleading or not and to what extent.