Frequentist Model Averaging with Applications
DescriptionThis project is an attempt to take the frequentist model averaging (FMA) literature a stepfurther from its current level that emphasises "general" statistical setups to one that focuseson "local" analysis and solutions. We consider estimation problems stemming from theanalysis of stochastic frontier and income inequality in Microeconometrics, and thesimulation of stochastic systems in Operations Research, where the choice of an appropriatefunctional form or distribution for the data, rather than the selection of covariates, is theprimary concern. Our objective is to develop an FMA strategy with optimal properties forestimating the parameter of interest in each case. The P.I. has explored the FMA strategiesthat can be usefully implemented for these problems, and already succeeded in deriving somekey theoretical properties of the methods. The requested amount is mainly for the recruitmentof a research associate/research fellow to carry out numerical and real data studies.
|Effective start/end date||1/10/16 → 21/05/19|