Model selection: An integral part of inference

摘要: We argue that model selection uncertainty should be fully incorporated into statistical inference whenever estimation is sensitive to choice and made with reference the data. consider different philosophies for achieving this goal suggest strategies data analysis. illustrate our methods through three examples. The first a Poisson regression of bird counts in which between inclusion one or both two covariates. second line transect set models yield substantially estimates abundance. third simulated example truth known.