AIC与BIC在亲体-补充量模型选择中的应用及比较

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。

基于AIC准则的锂离子电池变阶RC等效电路模型研究

提出一种变阶RC等效电路模型,并基于赤池信息量准则(AIC)辨识不同SOC处RC模型的最优阶数,兼顾了模型的准确度和实用性。通过脉冲充放电、恒流充放电以及自定义UDDS循环工况实验验证了该模型的有效性。变阶RC模型通过略增加模型的复杂度,能更加准确地描述锂离子电池两端陡、中间平的非线性电压特性,相对误差在1%以内,具有较高的实用价值。

基于AIC信息准则法的摆式列车倾摆伺服系统建模研究

通过对摆式列车倾摆控制系统输入、输出数据的研究,建立一个来自实验数据的倾摆控制系统模型。基于AIC信息准则法极大似然参数估计建模的基本思想和理论,分析摆式列车倾摆伺服系统的结构、作动原理及控制输入参考信号的特点,通过仿真获取摆式列车倾摆控制系统的输入、输出数据,用极大似然函数法对同类模型的不同模型结构进行参数估计,并以得到的估计参数计算出相应的似然函数值及AIC信息距离值,选取AIC信息距离最小的模型为倾摆伺服系统的模型。实验仿真结果表明:基于AIC信息准则的极大似然参数估计方法能够对系统模型进行建模估计,该模型不仅是同类模型中与实际系统误差最小的而且是最佳的。

AIC准则应用于动物最佳生长模型的选择

AIC准则作为选择回归模型的标准已有广泛应用，但在动物最佳生长模型的选择方面的应用情形尚未见报导。本文选择了几个经典的动物生长模型，把它们应用于描述家禽的生长过程，以AIC作为衡量模型优劣的标准去选择最佳生长模型。结果表明AIC准则应用于选择动物最佳生长模型方面是一种简单而行之有效的方法。

AIC准则与留一法交叉验证渐近等价的证明

AIC准则与留一法交叉验证是进行模型选择的常用方法。现有的证明二者渐近等价的方法是极大似然估计法。文章在线性模型的基础上,用删除残差与普通残差间的关系来推导AIC准则与留一法交叉验证对模型选择的效果,发现其是渐近等价的。这样的方法适用性更为广泛。最后通过仿真模拟和实际数据分析验证了结论。

基于SAIC方法的纵向数据模型平均

传统的SAIC模型平均所需运行时间随数据维数而呈现出阶乘级的增长,其预测精度也随之下降.本文基于传统SAIC模型平均法进行了改进,提出一类基于SAIC加权法的纵向数据模型平均法,使运算效率大幅提升,并且使预测效果拥有良好的稳定性.模拟实验结果表明,与传统方法相比,在预测残差平方和层面,本文提出的新模型在稳定性、精准性和运行速度方面均优于传统方法.

A primer on model selection using the Akaike Information Criterion

A powerful investigative tool in biology is to consider not a single mathematical model but a collection of models designed to explore different working hypotheses and select the best model in that collection.In these lecture notes,the usual workflow of the use of mathematical models to investigate a biological problem is described and the use of a collection of model is motivated.Models depend on parameters that must be estimated using observations;and when a collection of models is considered,the best model has then to be identified based on available observations.Hence,model calibration and selection,which are intrinsically linked,are essential steps of the workflow.Here,some procedures for model calibration and a criterion,the Akaike Information Criterion,of model selection based on experimental data are described.Rough derivation,practical technique of computation and use of this criterion are detailed.

SEMIPARAMETRIC MODEL SELECTION IN LARGE SAMPLES

For semiparametric regression model selection, based on a model selection criterion there is no finite order (or number of parameters) of the nonparametric part to be estimated consistently, but there is a finite order (or number of predictor variables) of the linear part to be estimated consistently. The models selected by using AIC and AICC are not consistent estimates of linear part of the true model. In this paper, we study the consistency in model selection by investigating the asymptotic properties of AIC* and AICC*, the modified versions of AIC and AICC respectively, which were proposed by a referee of the reference Shi and Tsai. Under some regular conditions, we prove that the parametric models of the semiparametric regression selected with AIC* and AICC* converge to the true model in probability. In addition, in terms of the mean integrated squared error plus a penalty, these two criteria can also provide an asymptotically efficient selection.

INFERENCE ON THE RANK OF THE GROWTH CURVE MODEL USING MODEL SELECTION METHOD

In this paper we study the estimation of the rank of the parameter trix(RPM) in a growth curve model in the framework of model selection. Following AIC criterion we propose a new general criterion and obtain a strongly consistent estimate of the RPM. We come to our conclusions under the assumptions of normal population and a general case separately.

Comparison of six statistical approaches in the selection of appropriate fish growth models

The performance of six statistical approaches,which can be used for selection of the best model to describe the growth of individual fish,was analyzed using simulated and real length-at-age data.The six approaches include coefficient of determination(R2),adjusted coefficient of determination(adj.-R2),root mean squared error(RMSE),Akaike's information criterion(AIC),bias correction of AIC(AICc) and Bayesian information criterion(BIC).The simulation data were generated by five growth models with different numbers of parameters.Four sets of real data were taken from the literature.The parameters in each of the five growth models were estimated using the maximum likelihood method under the assumption of the additive error structure for the data.The best supported model by the data was identified using each of the six approaches.The results show that R2 and RMSE have the same properties and perform worst.The sample size has an effect on the performance of adj.-R2,AIC,AICc and BIC.Adj.-R2 does better in small samples than in large samples.AIC is not suitable to use in small samples and tends to select more complex model when the sample size becomes large.AICc and BIC have best performance in small and large sample cases,respectively.Use of AICc or BIC is recommended for selection of fish growth model according to the size of the length-at-age data.

基于AIC的粗糙集择优方法

在实际应用中,当利用多种粗糙集构造算法所得到的多个粗糙集的误判率差异小时,误判率小的粗糙集并不总是具有最高预测准确度。利用粗糙集的分类规则构建Logistic模型,将拟合Logistic模型的AIC值作为该粗糙集的AIC值,用于粗糙集的择优。实例分析结果表明,采用新方法能够筛选出预测准确度较高的粗糙集。当多个粗糙集的误判率差异小时,新方法更可能选出预测准确度最高的粗糙集。

基于AIC准则的函数型数据主成分联合选择研究

函数型主成分分析(Functional Principal Component Analysis,FPCA)是对函数型数据进行降维的常用技术,本文将考虑函数型数据的主成分联合选择问题。首先,本文给出了两函数型变量的主成分联合模型,并通过基函数展开法和极大惩罚似然法对样本数据进行曲线平滑。在联合模型基础上,本文给出了确定函数型主成分个数的AIC准则,并提出了改进的ECME算法对模型参数进行估计。模拟显示AIC准则对应的主成分个数选择结果准确率更高,考虑两函数型数据之间相关信息的联合选择效果会比对各函数型数据主成分进行独立选择的结果有所提升。最后,本文将所提方法应用于老年人中医宗气数据的分析.

Plausible combinations: An improved method to evaluate the covariate structure of Cormack-Jolly-Seber mark-recapture models

Mark-recapture models are extensively used in quantitative population ecology, providing estimates of population vital rates, such as survival, that are difficult to obtain using other methods. Vital rates are commonly modeled as functions of explanatory covariates, adding considerable flexibility to mark-recapture models, but also increasing the subjectivity and complexity of the modeling process. Consequently, model selection and the evaluation of covariate structure remain critical aspects of mark-recapture modeling. The difficulties involved in model selection are compounded in Cormack-Jolly-Seber models because they are composed of separate sub-models for survival and recapture probabilities, which are conceptualized independently even though their parameters are not statistically independent. The construction of models as combinations of sub-models, together with multiple potential covariates, can lead to a large model set. Although desirable, estimation of the parameters of all models may not be feasible. Strategies to search a model space and base inference on a subset of all models exist and enjoy widespread use. However, even though the methods used to search a model space can be expected to influence parameter estimation, the assessment of covariate importance, and therefore the ecological interpretation of the modeling results, the performance of these strategies has received limited investigation. We present a new strategy for searching the space of a candidate set of Cormack-Jolly-Seber models and explore its performance relative to existing strategies using computer simulation. The new strategy provides an improved assessment of the importance of covariates and covariate combinations used to model survival and recapture probabilities, while requiring only a modest increase in the number of models on which inference is based in comparison to existing techniques.

Evaluation of numerical earthquake forecasting models

Evaluation of numerical earthquake forecasting models needs to consider two issues of equal importance:the application scenario of the simulation,and the complexity of the model.Criterion of the evaluation-based model selection faces some interesting problems in need of discussion.

Identifying the dependency pattern of daily rainfall of Dhaka station in Bangladesh using Markov chain and logistic regression model

Bangladesh is a subtropical monsoon climate characterized by wide seasonal variations in rainfall, moderately warm temperatures, and high humidity. Rainfall is the main source of irrigation water everywhere in the Bangladesh where the inhabitants derive their income primarily from farming. Stochastic rainfall models were concerned with the occurrence of wet day and depth of rainfall for different regions to model the daily occurrence of rainfall and achieved satisfactory results around the world. In connection to the Markov chain of different order, logistic regression is conducted to visualize the dependence of current rainfall upon the rainfall of previous two-time period. It had been shown that wet day of the previous two time period compared to the dry day of previous two time period influences positively the wet day of current time period, that is the dependency of dry-wet spell for the occurrence of rain in the rainy season from April to September in the study area. Daily data are collected from meteorological department of about 26 years on rainfall of Dhaka station during the period January 1985-August 2011 to conduct the study. The test result shows that the occurrence of rainfall follows a second order Markov chain and logistic regression also tells that dry followed by dry and wet followed by wet is more likely for the rainfall of Dhaka station and also the model could perform adequately for many applications of rainfall data satisfactorily.

Selecting the Quantity of Models in Mixture Regression

Mixture regression is a regression problem with mixed data. Specifically, in the observations, some data are from one model, while others from other models. Only after assuming the quantity of the model is given, EM or other algorithms can be used to solve this problem. We propose an information criterion for mixture regression model in this paper. Compared to ordinary information citizen by data simulations, results show our citizen has better performance on choosing the correct quantity of models.

A Model for the Mass-Growth of Wild-Caught Fish

The paper searched for raw data about wild-caught fish, where a sigmoidal growth function described the mass growth significantly better than non-sigmoidal functions. Specifically, von Bertalanffy’s sigmoidal growth function (metabolic exponent-pair a = 2/3, b = 1) was compared with unbounded linear growth and with bounded exponential growth using the Akaike information criterion. Thereby the maximum likelihood fits were compared, assuming a lognormal distribution of mass (i.e. a higher variance for heavier animals). Starting from 70+ size-at-age data, the paper focused on 15 data coming from large datasets. Of them, six data with 400 - 20,000 data-points were suitable for sigmoidal growth modeling. For these, a custom-made optimization tool identified the best fitting growth function from the general von Bertalanffy-Pütter class of models. This class generalizes the well-known models of Verhulst (logistic growth), Gompertz and von Bertalanffy. Whereas the best-fitting models varied widely, their exponent-pairs displayed a remarkable pattern, as their difference was close to 1/3 (example: von Bertalanffy exponent-pair). This defined a new class of models, for which the paper provided a biological motivation that relates growth to food consumption.

基于PCA-GEP算法的边坡稳定性预测

提出一种基于主成分分析的基因表达式程序设计算法,并将其用于边坡稳定性预测。该算法先采用主成分分析法对样本数据进行预处理,有效地减少预测模型的输入量,消除输入数据间的相关性,再将得到的新样本数据输入基因表达式,构建边坡稳定性的预测模型。利用该预测模型对82个危险圆弧破坏边坡实例中的71个实例进行学习,对另外11个实例进行预测,取得了较好的效果。在保留传统的以误差值作为评判模型优劣标准的同时,引入AIC信息准则法,分别对v-SVR算法和GA-BP网络算法和PCA-GEP算法三种预测模型进行比较分析,结果表明,运用该算法可以获得更优的预测模型,其预测结果比v-SVR算法和GA-BP网络等其他算法得到的结果具有更高的预测精度。工程实例计算表明,该方法是合理、可行的。

实证研究中预测模型的选择:从逐步回归到信息标准

本文首先对显著性变量同变量显著性之间的关系予以讨论并区分,进而评价逐步回归模型选择法的缺陷性。在此基础上,我们对以AIC和B IC为代表的各种基于信息标准的模型选择法予以介绍和评论。同逐步回归法相比,信息标准模型选择法有着坚实的统计理论基础及清晰而优良的统计性质。本文通过基于近十年中国股市数据的实证检验说明,信息标准同逐步回归相比往往能产生具有更强预测能力的计量模型,因此值得在未来的实证研究中注意并推广。

基于高斯模型的多重超声回波信号重数估计

精确估计多层材料超声回波信号的重数在超声检测上有着要意义。将小波变换方法用于多层材料超声回波参数估计中,根据高斯模型以超声回波信号的小波变换为基础、利用智能人工蜂群算法,估计出多重超声回波信号的各个参数。采用Akaike Information Criterion(AIC)准则,对叠加的两重和三重超声回波信号的重数进行估计。仿真结果表明,本算法可以实现多重超声回波信号重数的有效估计。用实验测试获得的回波对算法的性能进行了验证,结果证明了该算法的可行性和实用性。