Author:

Keywords:

Frequency domain, Granger causality, Gross domestic product, Predictive power, Stock prices

Abstract:

This paper investigates the predictive power for future domestic economic activity included in domestic stock prices, using a Granger causality analysis in the frequency domain. We are able to evaluate whether the predictive power is concentrated at the slowly fluctuating components or at the quickly fluctuating components. Using quarterly data for the G-7 countries, we found that the slowly fluctuating components of the stock prices have large predictive power for the future GDP, while this is not the case for the quickly fluctuating components. This finding holds both in a single-country setting and in a multi-country setting. Therefore, macro-economic policy makers could use the slowly fluctuating components of the stock prices to improve their predictions of the future GDP.