Model selection of copulas: AIC versus a cross validation copula information criterion

Akaike Information Criterion (AIC) is frequently employed in the semiparametric setting of selection of copula models, even though as a model selection tool it was developed in a parametric setting. Recently a Copula Information Criterion (CIC) has been especially designed for copula model selection. In this paper we examine the two approaches and present a simulation study where the performance of a cross-validated version of CIC is compared with the AIC criterion. Only minor differences are observed.     

Model selection using modified AIC and BIC in joint modeling of paired functional data

A modified version of the Akaike information criterion and two modified versions of the Bayesian information criterion are proposed to select the number of principal components and to choose the penalty parameters of penalized splines in a joint model of paired functional data. Numerical results show that, compared with an existing procedure using the cross-validation, the procedure based on the information criteria is computationally much faster while giving a similar performance.     

Generalizing cross redundancy in data envelopment analysis

Lee and Choi (2010) proved that a cross redundant output in a CCR or BCC DEA study is unnecessary and can be eliminated from the model without affecting the results of the study. A cross redundant output, as characterized by Lee and Choi, can be expressed as a specially constrained linear combination of both some outputs and some inputs. This article extends the contributions of Lee and Choi (2010) in at least three ways: (i) by adding precision and clarity to some of their definitions; (ii) by introducing specific definitions that complement the ones in their paper; and (iii) by conducting some additional analysis on the impact of the presence of other types of linear dependencies among the inputs and outputs of a DEA model. One reason that it is important to identify and remove cross redundant inputs or outputs from DEA models is that the computational burden of the DEA study is decreased, especially in large applications.     

On partitions classified by smallest missing part

The Ramanujan Journal - In this paper, we study partitions classified by its smallest missing part. Our main theorem is an inequality related to the parity of the smallest missing part of...     

Fast stepwise procedures of selection of variables by using AIC and BIC criteria

Fast stepwise procedures of selection of variables by using AIC and BIC criteria are proposed in this paper. We shall use a short name “FSP” for these new procedures. FSP are similar to the well-known stepwise regression procedures in computing steps. But FSP have two advantages. One of these advantages is that FSP are definitely convergent with a faster rate in finite computing steps. Another advantage is that FSP can be used for large number of candidate variables. In this paper we also show some asymptotic properties of FSP, and some simulation results.     

Optimality of AIC in Inference About Brownian Motion

In the usual Gaussian White-Noise model, we consider the problem of estimating the unknown square-integrable drift function of the standard Brownian motion using the partial sums of its Fourier series expansion generated by an orthonormal basis. Using the squared L ₂ distance loss, this problem is known to be the same as estimating the mean of an infinite dimensional random vector with l ₂ loss, where the coordinates are independently normally distributed with the unknown Fourier coefficients as the means and the same variance. In this modified version of the problem, we show that Akaike Information Criterion for model selection, followed by least squares estimation, attains the minimax rate of convergence. An erratum to this article can be found at     

A Bayesian analysis of the minimum AIC procedure

Summary By using a simple example a minimax type optimality of the minimum AIC procedure for the selection of models is demonstrated. The Institute of Statistical Mathematics     

Effective PCA for high-dimension, low-sample-size data with singular value decomposition of cross data matrix

In this paper, we propose a new methodology to deal with PCA in high-dimension, low-sample-size (HDLSS) data situations. We give an idea of estimating eigenvalues via singular values of a cross data matrix. We provide consistency properties of the eigenvalue estimation as well as its limiting distribution when the dimension d and the sample size n both grow to infinity in such a way that n is much lower than d. We apply the new methodology to estimating PC directions and PC scores in HDLSS data situations. We give an application of the findings in this paper to a mixture model to classify a dataset into two clusters. We demonstrate how the new methodology performs by using HDLSS data from a microarray study of prostate cancer.     

Goal programming approaches for data envelopment analysis cross efficiency evaluation

Cross efficiency evaluation has long been proposed as an alternative method for ranking the decision making units (DMUs) in data envelopment analysis (DEA). This study proposes goal programming models that could be used in the second stage of the cross evaluation. Proposed goal programming models have different efficiency concepts as classical DEA, minmax and minsum efficiency criteria. Numerical examples are provided to illustrate the applications of the proposed goal programming cross efficiency models.     

The Cauchy problem for semilinear hyperbolic equations with cross discontinuous data

This paper is devoted to study Cauchy problems of multidimensional semilinear strictly hyperbolic equations of second order with strongly singular initial data, where the derivatives of the initial data have discontinuity on two smooth curves transversally intersecting each other. The existence of the solution is proved, meanwhile, it is precisely discribed the flowery structure of the singularity of the solution.     

Bootstrapping Log Likelihood and EIC, an Extension of AIC

Akaike (1973, 2nd International Symposium on Information Theory, 267-281,Akademiai Kiado, Budapest) proposed AIC as an estimate of the expected loglikelihood to evaluate the goodness of models fitted to a given set of data.The introduction of AIC has greatly widened the range of application ofstatistical methods. However, its limit lies in the point that it can beapplied only to the cases where the parameter estimation are performed bythe maximum likelihood method. The derivation of AIC is based on theassessment of the effect of data fluctuation through the asymptoticnormality of MLE. In this paper we propose a new information criterion EICwhich is constructed by employing the bootstrap method to simulate the datafluctuation. The new information criterion, EIC, is regarded as an extensionof AIC. The performance of EIC is demonstrated by some numerical examples.     

Simultaneous estimation of means of classified normal observations

Simultaneous estimation of normal means is considered for observations which are classified into several groups. In a one-way classification case, it is shown that an adaptive shrinkage estimator dominates a Stein-type estimator which shrinks observations towards individual class averages as Stein's (1966,Festschrift for J. Neyman, (ed. F. N. David), 351–366, Wiley, New York) does, and is minimax even if class sizes are small. Simulation results under quadratic loss show that it is slightly better than Stein's (1966) if between variances are larger than within ones. Further this estimator is shown to improve on Stein's (1966) with respect to the Bayes risk. Our estimator is derived by assuming the means to have a one-way classification structure, consisting of three random terms of grand mean, class mean and residual. This technique can be applied to the case where observations are classified into a two-stage hierarchy.     

基于Spatial AIC准则的空间自回归模型变量选择研究

变量选择直接决定着空间计量经济模型的有效程度与实证研究结果。为有效解决空间自回归模型(即SAR模型)的变量选择问题,本文利用Kullback-Laible信息量最大化,把AIC准则运用到SAR模型构建,推导出Spatial AIC统计量,提出Spatial AIC准则。然后利用统计理论证明Spatial AIC准则选择SAR模型变量的渐近最优性;利用蒙特卡洛模拟方法,比较Spatial AIC准则、经典AIC准则和Lasso方法用于SAR模型变量选择的有限大样本性质;利用空间相关的沪深300成分股股票收益率数据,采用Spatial AIC准则和Lasso方法,分别构建股票收益率财务因素的空间自相关模型,实证比较其相对有效性。三种结果均表明Spatial AIC准则能够更好地解决SAR模型变量选择问题。     

Solving a large dense linear system by adaptive cross approximation

An efficient algorithm for the direct solution of a linear system associated with the discretization of boundary integral equations (in two dimensions) is described without having to compute the complete matrix of the linear system. This algorithm is based on the unitary-weight representation, for which a new construction based on adaptive cross approximation is proposed. This low rank approximation uses only a small part of the entries to construct the adaptive cross representation, and therefore the linear system can be solved efficiently.     

Analysis of rounded data from dependent sequences

Observations on continuous populations are often rounded when recorded due to the precision of the recording mechanism. However, classical statistical approaches have ignored the effect caused by the rounding errors. When the observations are independent and identically distributed, the exact maximum likelihood estimation (MLE) can be employed. However, if rounded data are from a dependent structure, the MLE of the parameters is difficult to calculate since the integral involved in the likelihood equation is intractable. This paper presents and examines a new approach to the parameter estimation, named as “short, overlapping series” (SOS), to deal with the α-mixing models in presence of rounding errors. We will establish the asymptotic properties of the SOS estimators when the innovations are normally distributed. Comparisons of this new approach with other existing techniques in the literature are also made by simulation with samples of moderate sizes.     

Analysis of the fiber length dependence of its strength by using the weakest-link approach 2. Analysis of test data

A method for estimating the model parameters is described, the results found in processing glass, flax, and carbon fiber and carbon bundle datasets are reported, and a comparison of different models is given. Translated from Mekhanika Kompozitnykh Materialov, Vol. 45, No. 1, pp. 65–74, January–February, 2008.