Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/244577 
Year of Publication: 
2021
Series/Report no.: 
Working Paper No. 2/2021
Publisher: 
Örebro University School of Business, Örebro
Abstract: 
We employ a battery of model evaluation tests for a broad-set of GARCH-MIDAS models and account for data snooping bias. We document that inferences based on standard tests for GM variance components can be misleading. Our data mining free results show that the gains of macro-variables in forecasting total (long run) variance by GM models are overstated (understated). Estimation of different components of volatility is crucial for designing differentiated investing strategies, risk management plans and pricing of derivative securities. Therefore, researchers and practitioners should be wary of data mining bias, which may contaminate a forecast that may appear statistically validated using robust evaluation tests.
Subjects: 
GARCH-MIDAS models
component variance forecasts
macro-variables
data snooping
JEL: 
C32
C52
G11
G17
Document Type: 
Working Paper

Files in This Item:
File
Size
850.35 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.