TY  - RPRT
TI  - Evaluation and selection of models for out-of-sample prediction when the sample size is small relative to the complexity of the data-generating process
AU  - Hannes Leeb
PY  - 2008
DO  - 10.3150/08-bej127
UR  - https://arxiv.org/abs/0802.3364
ID  - 0802.3364
ER  -