Title: Combining different frequencies in modeling non--negative processes: The MEM-MIDAS
Authors: Giampiero Gallo - NYU in Florence (Italy) [presenting]
Alessandra Amendola - University of Salerno (Italy)
Vincenzo Candila - University of Salerno (Italy)
Fabrizio Cipollini - University of Florence (Italy)
Abstract: In modeling financial time series, information of interest may be available at different frequency of observation. We extend the MIDAS-GARCH model to explicitly take in consideration that a multiplicative error model may be a more direct way to model the conditional expectation of a non-negative process observed daily and that a low frequency component in the data can be modeled exploiting some other information sampled at a different frequency, say monthly. The empirical application is presented on the realized volatility of the Dow Jones 30 components and the S\&P500.