AIC与BIC在亲体-补充量模型选择中的应用及比较

Comparision of AIC and BIC in the Selection of Stock-Recruitment Relationships

由于渔业资源评估中补充量的剧烈变动、亲体量的测量误差以及时间序列的偏差常常使亲体 补充量(SR)关系模型的确定存在很大偏差问题。本文以7种SR(Stock Recruitment)模型的模拟数据作为观测数据,研究了AIC(AkaikeInfor mationCriterion)与BIC(BayesianInformationCriterion)在SR模型选择中的应用。作为例证,文中采用AIC和BIC对8组实际的SR数据进行了SR模型的选择,并对其结果进行了比较。参数的估计方法为最大似然法(Maximumlikelihoodmethod)。结果表明,AIC和BIC在SR模型选择中是有效的。但是,对于嵌套模型,BIC可能比AIC更有效。

Variations in environmental variables and measurement errors often result in large and heterogeneous deviations in fitting fish stock-recruitment (SR) data to an SR statistical model. In this work, the maximum likelihood method was used to fit the seven statistical SR models on seven sets of simulated SR data. The best relationships were selected using AIC (Akaike Information Criterion) and BIC (Bayesian Information Criterion) methods, respectively, which has the advantage of testing the significance of the difference between the functions of different model specifications. The method was also utilized on eight sets of real fisheries SR data. The results showed that both AIC and BIC are valid in selecting the most suitable SR relationship. As far as the nested models are concerned, BIC is better than AIC.