作者: Peter Behl , Holger Dette , Manuel Frondel , Harald Tauchmann
DOI: 10.1016/J.ECONMOD.2011.09.002
关键词:
摘要: Abstract In contrast to conventional measures, the Focused Information Criterion (FIC) allows purpose-specific selection of models, thereby reflecting idea that one kind model might be appropriate for inferences on a parameter interest, but not another. Ever since its invention, FIC has been increasingly applied in realm statistics, this concept appears virtually unknown economic literature. Using straightforward analytical example, paper provides didactic illustration and demonstrates usefulness applications.