• Survey Research
• /
• v.12 no.3
• /
• pp.49-62
• /
• 2011

In this paper we investigate design-based properties of both the ordinary least square estimator and the weighted least square estimator for regression coefficients in panel regression model. We derive formulas of approximate bias, variance and mean square error for the ordinary least square estimator and approximate variance for the weighted least square estimator after linearization of least square estimators. Also we compare their magnitudes each other numerically through a simulation study. We consider a three years data of Korean Welfare Panel Study as a finite population and take household income as a dependent variable and choose 7 exploratory variables related household as independent variables in panel regression model. Then we calculate approximate bias, variance, mean square error for the ordinary least square estimator and approximate variance for the weighted least square estimator based on several sample sizes from 50 to 1,000 by 50. Through the simulation study we found some tendencies as follows. First, the mean square error of the ordinary least square estimator is getting larger than the variance of the weighted least square estimator as sample sizes increase. Next, the magnitude of mean square error of the ordinary least square estimator is depending on the magnitude of the bias of the estimator, which is large when the bias is large. Finally, with regard to approximate variance, variances of the ordinary least square estimator are smaller than those of the weighted least square estimator in many cases in the simulation.

• Journal of the Korea Safety Management & Science
• /
• v.10 no.3
• /
• pp.245-251
• /
• 2008

The research proposes three ANOVA(Analysis of Variance) tests using expected mean square(EMS) algorithms in various split-plot designs. The variance tests consist of Never-Pool test, Sometimes-Pool test and Always-Pool test. This paper also presents two EMS algorithms such as standard method and easy method. These algorithms are useful to make a decision rule for pooling. Numerical examples are illustrated for various split-plot designs such as split-plot designs, split-split-plot designs, repetition split-plot designs, and nested designs. Pragmatically, the results are summarized and compared with popular ANOVA spreadsheets and data model equations.

• The Korean Journal of Applied Statistics
• /
• v.22 no.2
• /
• pp.387-402
• /
• 2009

In this paper we have considered the problem of estimating the population mean $\bar{Y}$ of the study variable y using auxiliary information in presence of non-response. Classes of estimators for $\bar{Y}$ in the presence of non-response on the study variable y only and complete response on the auxiliary variable x is available, have been proposed in different situations viz., (i) population mean $\bar{X}$ is known, (ii) when population mean $\bar{X}$ and variance $S^2_x$ are known; (iii) when population mean $\bar{X}$ is not known: and (iv) when both population mean $\bar{X}$ and variance $S^2_x$ are not known: single and two-phase (or double) sampling. It has been shown that various estimators including usual unbiased estimator and the estimators reported by Rao (1986), Khare and Srivastava (1993, 1995) and Tabasum and Khan (2006) are members of the proposed classes of estimators. The optimum values of the first phase sample size n', second phase sample size n and the sub sampling fraction 1/k have been obtained for the fixed cost and the fixed precision. To illustrate foregoing, we have carried out an empirical investigation to reflect the relative performance of all the potentially competing estimators including the one due to Hansen and Hurwitz (1946) estimator, Rao (1986) estimator, Khare and Srivastava (1993, 1995) and Tabasum and Khan (2006) estimator.

• Industrial Engineering and Management Systems
• /
• v.16 no.1
• /
• pp.22-43
• /
• 2017

It is urgently-needed to construct a green supply chain (GSC) from collection of used products through recycling of them to sales of products using the recycled parts. Besides, it is necessary to consider the uncertainty in product demand as a risk in a GSC. This study proposes the optimal operations for a GSC with a retailer and a manufacturer. A retailer pays an incentive for collection of used products from customers and sells a single type of products in a market. A manufacturer produces the products ordered by the retailer, using recyclable parts with acceptable quality and compensates the collection cost of used products as to the recycled parts. This paper discusses the following risk attitudes: risk-neutral attitude, risk-averse attitude, and risk-prone attitude. Using mean-variance analysis, the optimal decisions for product order quantity, collection incentive, and lower limit of quality level, in the decentralized GSC (DGSC) and the integrated GSC (IGSC) are made. DGSC optimizes the utility function of each member. IGSC does that of the whole system. The analysis numerically investigates how (i) risk attitude and (ii) quality of recyclable parts affect the optimal operations. Supply chain coordination between GSC members to shift IGSC from DGSC is discussed.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.3
• /
• pp.805-809
• /
• 2006

In this paper a shrinkage estimator for the measure of dispersion of the inverse Gaussian distribution with known mean is proposed. Also we compare the relative bias and relative efficiency of the proposed estimator with respect to minimum variance unbiased estimator.

• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.5
• /
• pp.361-366
• /
• 2013

Prediction of process parameters is very important in parameter design. If predictions are fairly accurate, the quality improvement process will be useful to save time and reduce cost. The concept of dual response approach based on response surface methodology has widely been investigated. Dual response approach may take advantages of optimization modeling for finding optimum setting of input factor by separately modeling mean and variance responses. This study proposes an alternative dual response approach based on machine learning techniques instead of statistical analysis tools. A hybrid neural network-genetic algorithm has been proposed for the purpose of parameter design. A neural network is first constructed to model the relationship between responses and input factors. Mean and variance responses correspond to output nodes while input factors are used for input nodes. Using empirical process data, process parameters can be predicted without performing real experimentations. A genetic algorithm is then applied to find the optimum settings of input factors, where the neural network is used to evaluate the mean and variance response. A drug formulation example from pharmaceutical industry has been studied to demonstrate the procedures and applicability of the proposed approach.

• Korean Journal of Cognitive Science
• /
• v.10 no.3
• /
• pp.39-49
• /
• 1999

The space of IID([ndependently, Identically Distributed) textures was built with axes of mean luminance and contrast, and studied on what kind of mechanisms were required to mediate texture segregation in this space. The conjecture was tested that one of these mechanisms is sensitive to the differences between the means of textures to be discriminated, whereas the other is sensitive to the differences between variances. The probability of discrimination was measured for various pairs of textures in the lID space The data were well fit by a model in which discrimination depends on two mechanisms whose responses are combined by probability summation. The conjecture was rejected that two mechanisms respectively tuned to mean and variance of texture function in segregation. Discrimination within space is mediated by 2 independent channels however: the 2 independent channels are not exactly tuned to texture mean and variance. One m mechanism was primarily sensitive to texture mean, whereas the other was sensitive to b both texture mean and variance.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.3
• /
• pp.381-387
• /
• 2011

This paper suggests a method for estimating components of variance in one-factor random model. Estimates of variance components are given by the method of moments. Sums of squares due to variance sources are obtained by projections. This paper also shows how to use eigenvalues for getting the coefficients of variance components in the expression of the expectations of the mean squares. The suggested method shows easier and faster than the method of Harley's synthesis.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.1
• /
• pp.80-89
• /
• 2013

This paper presents an efficient approach for solving economic dispatch (ED) problems with nonconvex cost functions using a 'Mean-Variance Optimization (MVO)' algorithm with Kuhn-Tucker condition and swap process. The aim of the ED problem, one of the most important activities in power system operation and planning, is to determine the optimal combination of power outputs of all generating units so as to meet the required load demand at minimum operating cost while satisfying system equality and inequality constraints. This paper applies Kuhn-Tucker condition and swap process to a MVO algorithm to improve a global minimum searching capability. The proposed MVO is applied to three different nonconvex ED problems with valve-point effects, prohibited operating zones, transmission network losses, and multi-fuels with valve-point effects. Additionally, it is applied to the large-scale power system of Korea. The results are compared with those of the state-of-the-art methods as well.

• Communications for Statistical Applications and Methods
• /
• v.20 no.4
• /
• pp.329-337
• /
• 2013

Consider a p-variate($p{\geq}3$) normal distribution with mean ${\theta}$ and covariance matrix ${\sum}={\sigma}^2{\mathbf{I}}_p$ for any unknown scalar ${\sigma}^2$. In this paper we improve the James-Stein estimator of ${\theta}$ in cases of shrinking toward some vectors using the Stein variance estimator. It is also shown that this domination does not hold for the positive part versions of these estimators.