Testing for wavelet based time-frequency relationship between oil prices and US economic activity

Syed Ali Raza , Muhammad Shahbaz (1) , Rafi Amir-Ud-Din , Rashid Sbia (2) , Nida Shah

1 GATECH - College of Computing
2 AMSE - Aix-Marseille Sciences Economiques

This study investigates the empirical association of oil prices with economic activity in developed open economy namely: The United States by using the wavelet transform framework. This methodology enables the decomposition of time-series at different time-frequencies. In this study, we have used maximal overlap discrete wavelet transform, wavelet covariance, wavelet correlation, continuous wavelet power spectrum, wavelet coherence spectrum and wavelet based Granger causality approaches to analyze the relationship between oil prices and economic activity. The present study uses month frequency data for the period of 1979M1-2013M7. The results indicate that oil prices have positive impact on economic activity and the feedback effect exists between oil prices and economic activity.

Mots clés: Discrete wavelet analysis; Wavelet coherence; Oil prices; Economic activity

Domaines: Economies et finances

Syed Ali Raza, Muhammad Shahbaz, Rafi Amir-Ud-Din, Rashid Sbia, Nida Shah. Testing for wavelet based time-frequency relationship between oil prices and US economic activity. Energy, 2018, 154, pp.571-580. ⟨10.1016/j.energy.2018.02.037⟩. ⟨hal-01982294⟩