• Communications for Statistical Applications and Methods
• /
• v.7 no.2
• /
• pp.617-631
• /
• 2000

This paper is concerned with suggesting a Bayesian method for variable selection in generalized logit model. It is based on Laplace-Metropolis algorithm intended to propose a simple method for estimating the marginal likelihood of the model. The algorithm then leads to a criterion for the selection of variables. The criterion is to find a subset of variables that maximizes the marginal likelihood of the model and it is seen to be a Bayes rule in a sense that it minimizes the risk of the variable selection under 0-1 loss function. Based upon two examples, the suggested method is illustrated and compared with existing frequentist methods.

• Proceedings of the Korea Association of Information Systems Conference
• /
• 2002.11a
• /
• pp.289-312
• /
• 2002

This study suggests the Link weight analysis approach to choose input variables and an integrated model to make more accurate bankruptcy prediction model. the Link weight analysis approach is a method to choose input variables to analyze each input node's link weight which is the absolute value of link weight between an input nodes and a hidden layer. There are the weak-linked neurons elimination method, the strong-linked neurons selection method in the link weight analysis approach. The Integrated Model is a combined type adapting Bagging method that uses the average value of the four models, the optimal weak-linked-neurons elimination method, optimal strong-linked neurons selection method, decision-making tree model, and MDA. As a result, the methods suggested in this study - the optimal strong-linked neurons selection method, the optimal weak-linked neurons elimination method, and the integrated model - show much higher accuracy than MDA and decision making tree model. Especially the integrated model shows much higher accuracy than MDA and decision making tree model and shows slightly higher accuracy than the optimal weak-linked neurons elimination method and the optimal strong-linked neurons selection method.

• The Korean Journal of Applied Statistics
• /
• v.16 no.2
• /
• pp.335-349
• /
• 2003

In this paper, we use Bayesian method for model selection of poisson vs. negative binomial distribution, and normal, double exponential vs. cauchy distribution. The fractional Bayes factor of O'Hagan (1995) was applied to Bayesian model selection under the assumption of noninformative improper priors for all parameters in the models. Through the analyses of real data and simulation data, we examine the usefulness of the fractional Bayes factor in comparison with intrinsic Bayes factors of Berger and Pericchi (1996, 1998).

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.5
• /
• pp.1317-1325
• /
• 2016

The purpose of this paper is to explain the factors affecting the wage of the vocational high school graduates. We particularly examine the effectiveness of controlling sample selection bias by employing the Tobit model and Heckman sample selection model. The major results are as follows. First it is shown that the Tobit model and Heckman sample selection model controlling sample selection bias is statistically significant. Hence all the independent variables seem to be statistically consistent with the theoretical model. Second, gender was statistically significant, both in the probability of employment and the wage. Third, the employment probability and wage of Maester high school graduates were shown to be high compared to all other graduates. Fourth, the higher parent's income, the higher are both the employment probability and the wage. Finally, parents education level, high school grade, satisfaction, and a number of licenses were found to be statistically significant, both in the probability of employment and wages.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.6
• /
• pp.1191-1197
• /
• 2009

Most Bayesian approaches to model selection of wavelet analysis have drawbacks that computational cost is expensive to obtain accuracy for the fitted unknown function. To overcome the drawback, this article introduces loss functions which are criteria for level dependent threshold selection in wavelet based Bayesian methods with arbitrary size and regular design points. We demonstrate the utility of these criteria by four test functions and real data.

• Journal of Information Processing Systems
• /
• v.2 no.2
• /
• pp.130-134
• /
• 2006

In the distributed replicating server model, the provision of replicated services will improve the performance of the providing service and efficiency for clients. Efficiently composing the server selection algorithm decreases the retrieval time for replicated data. In this paper, we define the system model that selects and connects the replicated server that provides an optimal service using the server-side downstream measurement and propose a server selection algorithm.

• Journal of the Korean Statistical Society
• /
• v.27 no.4
• /
• pp.397-405
• /
• 1998

We develop parametric bootstrap model selection criteria in an example to fit a random sample to either a general normal distribution or a normal distribution with prespecified mean. We apply the bootstrap methods in two ways; one considers the direct substitution of estimated parameter for the unknown parameter, and the other focuses on the bias correction. These bootstrap model selection criteria are compared with AIC. We illustrate that all the selection rules reduce to the one sample t-test, where the cutoff points converge to some certain points as the sample size increases.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.313-316
• /
• 2004

Model selection is the process that sets up the regularization parameter in the support vector machine or regularization network by using the external methods such as general cross validation or L-curve criterion. This paper suggests that the regularization parameter can be obtained simultaneously within the learning process of neural networks without resort to separate selection methods. In this paper, extended kernel method is introduced. The relationship between regularization parameter and the bias term in the extended kernel is established. Experimental results show the effectiveness of the new model selection method.

• The Korean Journal of Applied Statistics
• /
• v.29 no.4
• /
• pp.643-656
• /
• 2016

High dimensional data are frequently encountered in various fields where the number of variables is greater than the number of samples. It is usually necessary to select variables to estimate regression coefficients and avoid overfitting in high dimensional data. A penalized regression model simultaneously obtains variable selection and estimation of coefficients which makes them frequently used for high dimensional data. However, the penalized regression model also needs to select the optimal model by choosing a tuning parameter based on the model selection criterion. This study deals with the bias effect of LASSO regression for model selection criteria. We numerically describes the bias effect to the model selection criteria and apply the proposed correction to the identification of biomarkers for lung cancer based on gene expression data.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.3
• /
• pp.613-618
• /
• 2011

The proposed method is based on a penalized log partial likelihood of Cox proportional hazard model with L1-penalty. We use the iteratively reweighted least squares procedure to solve L1 penalized log partial likelihood function of Cox proportional hazard model. It provide the ecient computation including variable selection and leads to the generalized cross validation function for the model selection. Experimental results are then presented to indicate the performance of the proposed procedure.