• Journal of the Korean Statistical Society
• /
• v.31 no.1
• /
• pp.63-75
• /
• 2002

Most attempts at Bayesian analysis of neural networks involve hierarchical modeling. We believe that similar results can be obtained with simpler models that require less computational effort, as long as appropriate restrictions are placed on parameters in order to ensure propriety of posterior distributions. In particular, we adopt a model first introduced by Lee (1999) that utilizes an improper prior for all parameters. Straightforward Gibbs sampling is possible, with the exception of the bias parameters, which are embedded in nonlinear sigmoidal functions. In addition to the problems posed by nonlinearity, direct sampling from the posterior distributions of the bias parameters is compounded due to the duplication of hidden nodes, which is a source of multimodality. In this regard, we focus on sampling from the marginal posterior distribution of the bias parameters with Markov chain Monte Carlo methods that combine traditional Metropolis sampling with a slice sampler described by Neal (1997, 2001). The methods are illustrated with data examples that are largely confined to the analysis of nonparametric regression models.

• Journal of Environmental Impact Assessment
• /
• v.6 no.1
• /
• pp.111-119
• /
• 1997

This study was carried out to suggest the best statistical test which may be used to quantify the change of water quality between two groups. Traditional t-test may not be used in cases where the normality of underlying population distribution is not assured. Three non-parametric tests which are based on the relative order of the measurements, were studied to find out the applicability in water quality data analysis. The sign test is based on the sign of the deviation of the measurement from the median value, and the binomial distribution table is used. The signed rank test utilizes not only the sign but also the magnitude of the deviation. The Wilcoxon rank-sum test which is basically same as Mann-Whitney test, tests the mean difference between two independent samples which may have missing data. Among the three non-parametric tests studied, the singed rank test was found out to be applicable in the quantification of the change of water quality between two samples.

• Journal of Periodontal and Implant Science
• /
• v.45 no.1
• /
• pp.2-7
• /
• 2015

A fundamental problem in analyzing complex multilevel-structured periodontal data is the violation of independency among the observations, which is an assumption in traditional statistical models (e.g., analysis of variance and ordinary least squares regression). In many cases, aggregation (i.e., mean or sum scores) has been employed to overcome this problem. However, the aggregation approach still exhibits certain limitations, such as a loss of power and detailed information, no cross-level relationship analysis, and the potential for creating an ecological fallacy. In order to handle multilevel-structured data appropriately, mixed effects models have been introduced and employed in dental research using periodontal data. The use of mixed effects models might account for the potential bias due to the violation of the independency assumption as well as provide accurate estimates.

• Korean Journal of Child Studies
• /
• v.27 no.3
• /
• pp.169-187
• /
• 2006

A hierarchical linear model(HLM) provides advantages over existing traditional statistical methods (e.g., ordinary least squares regression, repeated measures analysis of variance, etc.) for analyzing multilevel/longitudinal data or diary methods. HLM can gauge a more precise estimation of lower-level effects within higher-level units, as well as describe each individual's growth trajectory across time with improved estimation. This article 1) provides scholars who study children and families with an overview of HLM (i.e., statistical assumptions, advantages/disadvantages, etc.), 2) provides an empirical study to illustrate the application of HLM, and 3) discusses the application of HLM to the study of children and families. In addition, this article provided useful information on available articles and websites to enhance the reader's understanding of HLM.

• Journal of Information Technology Applications and Management
• /
• v.24 no.3
• /
• pp.93-105
• /
• 2017

Demands of cosmetics are affected not only by the well-known elements such as brand, price, and customer's consumption capacity, but also by some latent factors, for example, weather and air environment. Due to complexity and dynamic changes of the above factors, their influences can hardly be estimated in an accurate way by the traditional approaches such as survey and questionnaires. Through modeling and statistical analysis of big data, this article studied the impacts of weather condition and air quality on customer flow and sales of the cosmetics distributors in China, and found several hidden influencing factors. It provided a big-data based method for the analysis of unconventional factors on cosmetics marketing in the changing weather condition and air environment.

• Food Science and Biotechnology
• /
• v.15 no.4
• /
• pp.494-498
• /
• 2006

An increase in variance (overdispersion) can occur when a binomial statistical analysis is applied to sensory difference test data in which replicate sensory evaluations (tastings) and multiple evaluators (judges) are combined to increase the sample size. Such a practice can cause extensive Type I errors, leading to serious misinterpretations of the data, especially when traditional simple binomial analysis is applied. Alternatively, the use of beta binomial analysis will circumvent the problem of overdispersion. This brief review discusses the uses and computation methodology of beta binomial analysis and in practice evidence for the occurrence of overdispersion.

• The Journal of Fisheries Business Administration
• /
• v.52 no.1
• /
• pp.23-46
• /
• 2021

The purpose of this study is to understand the production efficiency of individual fishing communities and provide directions for improvement. The subject of the study is aquaculture type Ochon-Gye in Goheung-gun. The analysis method used bootstrap-DEA to overcome the statistical reliability problem of the traditional DEA analysis technique. In addition, data mining-GIS was applied to identify the spatial productivity of fishing communities. The values of technology efficiency, pure technology efficiency, and scale efficiency were estimated for 32 aquaculture-type fishing villages. Then, using the benchmarking reference set and weights, the projection was presented through adjustment of the input factor excess, and furthermore, the confidence interval of the efficiency values considering statistical significance was estimated using bootstrap.

• Communications for Statistical Applications and Methods
• /
• v.25 no.6
• /
• pp.633-645
• /
• 2018

Gaussian error distributions are a common choice in traditional regression models for the maximum likelihood (ML) method. However, this distributional assumption is often suspicious especially when the error distribution is skewed or has heavy tails. In both cases, the ML method under normality could break down or lose efficiency. In this paper, we consider the log-concave and Gaussian scale mixture distributions for error distributions. For the log-concave errors, we propose to use a smoothed maximum likelihood estimator for stable and faster computation. Based on this, we perform comparative simulation studies to see the performance of coefficient estimates under normal, Gaussian scale mixture, and log-concave errors. In addition, we also consider real data analysis using Stack loss plant data and Korean labor and income panel data.

• Communications for Statistical Applications and Methods
• /
• v.26 no.1
• /
• pp.57-67
• /
• 2019

Income distribution is a major concern in economic theory. In regional economics, it is often of interest to compare income distributions in different regions. Traditional methods often compare the income inequality of different regions by assuming parametric forms of the income distributions, or using summary statistics like the Gini coefficient. In this paper, we propose a nonparametric procedure to test for heterogeneity in income distributions among different regions, and a K-means clustering procedure for clustering income distributions based on energy distance. In simulation studies, it is shown that the energy distance based method has competitive results with other common methods in hypothesis testing, and the energy distance based clustering method performs well in the clustering problem. The proposed approaches are applied in analyzing data from China Health and Nutrition Survey 2011. The results indicate that there are significant differences among income distributions of the 12 provinces in the dataset. After applying a 4-means clustering algorithm, we obtained the clustering results of the income distributions in the 12 provinces.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.6
• /
• pp.2893-2907
• /
• 2019

Wireless mobile communication systems in subway tunnels have been widely researched these years, due to increased demand for the communication applications. As a result, an accurate model is essential to effectively evaluate the communication system performance. Thus, a neoteric three-dimensional (3D) geometry-based stochastic model (GBSM) is proposed for the massive multiple-input multiple-output (MIMO) fading channels in tunnel environment. Furthermore, the statistical properties of the channel such as space-time correlation, amplitude and phase probability density are analyzed and compared with those of the traditional two-dimensional (2D) model by numerical simulations. Finally, the ergodic capacity is investigated based on the proposed model. Numerical results show that the proposed model can describe the channel in tunnels more practically.