Is AIC the right criterion?
- For small data and frequentist inference, you should use AICc -- the small sample correction which provides greater penalty for each parameter but approaches AIC as n becomes large. If it makes a difference, you should use it. (I probably should have used it above.)
- For large data and frequentist inference , consider BIC, which is asymptotically consistent while AIC is not (see Hastie et al. 2009, which is available online). AIC typically favors overly-complex models with large n relative to BIC. Note, however, that this is not an issue for prediction, only inference of a true model (if one exists; Sec. 6.4 McElreath 2015).
- For Bayesian models, consider WAIC or LOO (instead of DIC, which has issues with non-Gaussian posteriors McElreath 2015).
- Don't use information criteria for model selection between GLMs with different link functions.
AIC是正确的标准吗？
- 对于小型数据和频率论推断，你应该使用AIC_C -小样本校正功能，而且提供了更大的惩罚每一个参数，但随着n变大方法AIC。 如果它有区别，你应该使用它。 （我可能应该使用它上面。）
- 对于大数据和频率论推断，考虑BIC，这是渐近一致的，而AIC不是（见黑斯蒂等人。2009年，也就是在网上提供）。 AIC通常主张相对于BIC大的N过于复杂的模型。 但是请注意，这不是对预测的问题，只有一个真正的模型的推理（如果存在;二段6.4 McElreath 2015）。
- 对于贝叶斯模型，考虑WAIC或LOO （而不是DIC，DIC有非高斯后验问题 McElreath 2015年）。
- 不要在具有不同链接函数的GLM之间使用信息标准进行模型选择。
