Performances of LOO and WAIC as IRT Model Selection Methods

@inproceedings{Luo2017PerformancesOL,
  title={Performances of LOO and WAIC as IRT Model Selection Methods},
  author={Yong Luo and Khaleel A. Al-harbi},
  year={2017}
}
The widely available information criterion (WAIC) and leave-one-out cross-validation (LOO) are considered fully Bayesian model selection methods due to their utilization of the whole posterior distribution other than the point estimates. Despite their theoretical advantage of being fully Bayesian, how such an advantage translates into practical performance remains unknown. In this paper, we conducted a simulation study to compare the performances of WAIC and LOO with other four commonly used… CONTINUE READING

Figures and Tables from this paper.

Similar Papers