Reliable Statistical Inferences with Wrong Probability Models (White, 1982, Econometrica; White, 1994, Estimation, Inference, and Specification Analysis;see Golden, 1995, Journal of Mathematical Psychology; for a review; Golden, 2000)
Classical statistical inference (assumes DGP ? model):
- ?n estimates true parameter vector ?*.
- Wrong inferences for wrong models (“true parameter” doesn’t exit)
DRMST Theory (assumes weak constaints on DGP):
- ?n estimates global minimum ?* of objective function d?
- Global minimum ? “true” parameter vector ?* (if DGP ? model)
DRMST Theory disentangles issues of: Reliable inference and Valid inference.
Philosophy: Since any model is “imperfect” ? “Classical” approach is not appropriate