Frequency Domain Analysis of Consumer Confidence and Industrial Production in Turkey

This paper examines the relationship between consumer confidence and economic growth for the Turkish economy employing frequency domain analysis. We use a monthly data set with industrial production (as proxy for economic growth) and CNBC-e consumer confidence index covering the period 2002:01-2009:12. Our methodology includes the test developed by Breitung and Candelon (2006) as well as conventional tools of spectrum analysis such as cospectrum, squared coherence and phase and gain spectrums. We focus on decomposing total covariance between consumer confidence and industrial production across main frequency bands as well as attainimg lead/lag interactions between them. Our preliminary empirical findings show that variations in consumer confidence mainly concentrate over seasonal frequencies. Nevertheless, we also observe significant feedbacks from economic growth to short-term consumer confidence over high frequencies. On the other hand, spectral variance decompositions show that percentage of variation in consumer confidence due to economic growth is much higher than the percentage of variation in economic growth due to consumer confidence.