作者: Kenneth P. Burnham , David R. Anderson
DOI: 10.1007/978-94-011-2868-1_3
关键词: Model selection 、 Selection (genetic algorithm) 、 Inference 、 Computer science 、 Basis (linear algebra) 、 Artificial intelligence 、 Estimator 、 Machine learning 、 Key (cryptography) 、 Occam's razor 、 Structure (mathematical logic)
摘要: Selection of an appropriate model as the basis for data analysis is critical valid inference. Fundamental to this issue concept that datawill only “support” limited A should have enough structure and parameters account adequately significant variability in data, otherwise bias estimators likely. However, if has too much or many parameters, then precision unnecessarily lost “effects” may be inferred are not justified by data. proper fully supported avoid bias, but (the Principle Parsimony) .Thus, given there a need choose objectively from among alternative models, each based on biological considerations.