Volatility Clustering in Monthly Stock Returns

We investigate volatility clustering using a modeling approach based on the temporal aggregation results for generalized autoregressive conditional heteroscedasticity (GARCH) models in Drost and Nijman [Econometrica, 1993]. Our findings highlight that volatility clustering, contrary to widespread belief, is not only present in high-frequency financial data. Monthly data also exhibit significant serial dependence in the second moments. We show that the use of temporal aggregation to estimate low-frequency models reduces parameter uncertainty substantially