Data-Based Selection of an Appropriate Biological Model: The Key to Modern Data Analysis

作者: Kenneth P. Burnham , David R. Anderson

DOI: 10.1007/978-94-011-2868-1_3

关键词: Model selectionSelection (genetic algorithm)InferenceComputer scienceBasis (linear algebra)Artificial intelligenceEstimatorMachine learningKey (cryptography)Occam's razorStructure (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.

参考文章(30)
Statistical data analysis American Mathematical Society. ,vol. 28, ,(1983) , 10.1090/PSAPM/028
Ritei Shibata, Statistical aspects of model selection From data to model. pp. 215- 240 ,(1989) , 10.1007/978-3-642-75007-6_5
Cohn L. Mallows, More Comments onCp Technometrics. ,vol. 37, pp. 362- 372 ,(1995) , 10.1080/00401706.1995.10484370
Hirotugu Akaike, Prediction and Entropy Springer, New York, NY. pp. 1- 24 ,(1985) , 10.1007/978-1-4613-8560-8_1
Peter McCullagh, John Ashworth Nelder, Generalized Linear Models ,(1983)
Gary C. White, Numerical estimation of survival rates from band-recovery and biotelemetry data Journal of Wildlife Management. ,vol. 47, pp. 716- ,(1983) , 10.2307/3808607
D. R. COX, Some remarks on overdispersion Biometrika. ,vol. 70, pp. 269- 274 ,(1983) , 10.1093/BIOMET/70.1.269