Abstract—This study presents a Bayesian regression method for estimating the dynamic dependence of the stationary component of Gross Domestic Product (GDP) on that of capacity utilization. First, stationary components from original GDP and capacity utilization time series are extracted using a set of state space models. Then, a set of Bayesian regression models with a time-varying coefficient is constructed. As an application of our proposed method, we analyze the dynamic relationship between GDP and capacity utilization using Japanese economy data from 1980 to 2005. The results imply that capacity utilization has had a larger influence on GDP in the expansion phase than in the recession phase of the Japanese business cycle since the 1990s.

Index Terms—Bayesian modeling, dynamic relationship analysis, time-varying coefficient, gross domestic product, capacity utilization rate, Japanese economy.

Cite: Koki Kyo and Hideo Noda, "Dynamic Relationship between Business Fluctuation and Capacity Utilization in the Japanese Manufacturing Industry," Journal of Industrial and Intelligent Information, Vol. 6, No. 2, pp. 57-62, December 2018. doi: 10.18178/jiii.6.2.57-62