Shaping the manufacturing industry performance: MIDAS approach

We aim to find out whether the exchange rate (against US dollar) or the interest rate (in local currency) is a better variable in predicting the capacity utilization rate of manufacturing industry (CUR) of Turkey after the 2008 global financial crisis. In that manner, we implement dynamic mixed data sampling (MIDAS) regression model to forecast monthly changes in CUR by using daily changes in the exchange rate and the interest rate separately. The results show that exchange rate has a better forecast performance suggesting that it is a stronger determinant in shaping the manufacturing industry.