• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2005년도 추계 학술발표회 논문집
• /
• pp.181-186
• /
• 2005

Many data sets obtained from surveys or medical trials often include missing observations. When these data sets are analyzed, it is general to use only complete cases. However, it is possible to have big biases or involve inefficiency. In this paper, we consider a method for estimating parameters in logistic linear models involving non-ignorable missing data mechanism. A binomial response and normal exploratory model for the missing data are used. We fit the model using the EM algorithm. The E-step is derived by Metropolis-hastings algorithm to generate a sample for missing data and Monte-carlo technique, and the M-step is by Newton-Raphson to maximize likelihood function. Asymptotic variances of the MLE's are derived and the standard error and estimates of parameters are compared.

• Communications for Statistical Applications and Methods
• /
• 제9권1호
• /
• pp.229-240
• /
• 2002

Whether to use linear or quadratic model in the analysis of regression data is one of the important problems in classical regression model and measurement error model (MEM). In MEM, four goodness of fit test statistics are available In solving that problem. Two are from the derivation of estimators of quadratic MEM, and one is from that of the general $k^{th}$-order polynomial MEM. The fourth one is derived as a variation of goodness of fit test statistic used in linear MEM. The purpose of this paper is to find the most powerful test statistic among them through the small-scale simulation.

• Journal of the Korean Data and Information Science Society
• /
• 제15권1호
• /
• pp.251-261
• /
• 2004

The paper focuses primarily on the standard linear multiple regression model where the parameter of interest is a ratio of two regression coefficients. The general model includes the calibration model, the Fieller-Creasy problem, slope-ratio assays, parallel-line assays, and bioequivalence. We provide an orthogonal transformation (cf. Cox and Reid (1987)) of the original parameter vector. Also, we give some remarks on the difficulties associated with likelihood based confidence interval.

• Journal of applied mathematics & informatics
• /
• 제41권1호
• /
• pp.215-228
• /
• 2023

The ready-made garments (RMG) have been making a crucial contribution about of 81% of total export and 12.36 % of total GDP of the country which is now the single biggest export earner for Bangladesh. The cheap production cost is the key important factor to explore this RMG sector. But these RMG sector is running on the basis of intuition based decisions. Though they are making profit it is not optimal. In this study, a deterministic model is developed to help the RMG to minimize the production cost and to maximize their profit along with optimal utilization of available resources. 10 different types of products are taken from one of the garments factories of Gazipur, Dhaka to prepare this research work. This model suggests the manufacturer on which products along with how much should be produced to meet the future demand by maintaining the lowest production cost that ultimately maximize the profit of the organization, and also helps Bangladesh to compete in the international market with 'Made in Bangladesh'. LINDO programming is used here to solve this LP model.

• 상하수도학회지
• /
• 제13권1호
• /
• pp.95-112
• /
• 1999

In this research, it is proposed that optimization method is introduced and applied to the design of pipeline system in multi-regional water supply project, which has been constructed to settle the regional unbalance problems of available water resources. For the purpose, interface programs are developed to integrate linear programming model and KYPIPE model which is used for optimization and hydraulic analysis, respectively. The developed program is applied to the pipeline system design of multi-regional water supply project. The optimal diameters from the application of linear programming technique are compared with those from conventional method that is time-consuming and tedious trail and error process. Since the conventional design largely depends upon the experience of designers and the results of general hydraulic analysis, it can not be reasonable and consistent. The application of linear programming technique can make it possible to design pipeline system optimally by using same design factors of general hydraulic models. The model can select commercial discrete pipe diameter as optimal size by using pipe length as decision variables. The developed model is applied to Pohang multi-regional water supply system design with two different objective functions, which are initial construction cost and annual cost including electric cost. As results, it is calculated that the initial construction cost of 1,449,740 thousand won is saved and annual cost of 128,951 thousand won is saved for a year within study year. Also, the optimal site of pump station is selected on 5th pipe, which is located between the diverging junction to Kangdong(2) province and the diverging junction to Cheonbuk province. It is explained that pump cost is less than pipe cost in this application case study due to little pump station scale. In the case of water supply with large pump capacity, it is reasonal that the increase of pipe size is more efficient instead the increase of pump station capacity to save annual cost.

• 한국경영과학회지
• /
• 제11권1호
• /
• pp.36-43
• /
• 1986

In this study, the fuzzy programming is extended to handle various types of membership functions by transformation of the complicated fuzzy programming problems into the equivalent crisp linear programming problems with single objective. It is well-known that the fuzzy programming problem with linear membership functions (i.e., ramp type) can be easily transformed into a linear programming problem by introducing one dummy variable to minimize the worst unwanted deviation. However, until recently not many researches have been done to handle various general types of complicated linear membership functions which might be more realistic than ramp-or triangular-type functions. In order to handle these complicated membership functions, the goal dividing concept, which is based on the fuzzy set operation (i. e., intersection and union operations), has been prepared. The linear model obtained using the goal dividing concept is more efficient and single than the previous models [4, 8]. In addition, this result can be easily applied to any nonlinear membership functions by piecewise approximation since the membership function is continuous and monotone increasing or decreasing.

• 전기학회논문지
• /
• 제63권4호
• /
• pp.534-540
• /
• 2014

In this paper, we propose a fuzzy combined Polynomial Neural Network(PNN) for pattern classification. The fuzzy combined PNN comes from the generic TSK fuzzy model with several linear polynomial as the consequent part and is the expanded version of the fuzzy model. The proposed pattern classifier has the polynomial neural networks as the consequent part, instead of the general linear polynomial. PNNs are implemented by stacking the simple polynomials dynamically. To implement one layer of PNNs, the various types of simple polynomials are used so that PNNs have flexibility and versatility. Although the structural complexity of the implemented PNNs is high, the PNNs become a high order-multi input polynomial finally. To estimate the coefficients of a polynomial neuron, The weighted linear discriminant analysis. The output of fuzzy rule system with PNNs as the consequent part is the linear combination of the output of several PNNs. To evaluate the classification ability of the proposed pattern classifier, we make some experiments with several machine learning data sets.

• 대한기계학회논문집B
• /
• 제30권7호
• /
• pp.612-621
• /
• 2006

Drawing is a mechanical operation attenuating material thickness to an appropriate level for the next processing or end usage. When the input material has a form of bundle or bundles made of very thin and long shaped wires or fibers, this attenuation operation is called 'bundle drawing' or 'drafting'. Bundle drawing is being used widely in manufacturing micro sized wires or staple yarns. However, the bundle processed by this operation has more or less defects in the evenness of linear density. Such irregularities cause many problems not only for the product quality but also for the efficiency of the next successive processes. In this research a mathematical model for the dynamic behavior of the bundle fluid is to be set up on the basis of general physical laws containing physical variables, i.e. linear density and velocity as the dynamic state variables of the bundle fluid. The governing equations resulting from the modeling show that they appear in a slightly different form from what they do in a continuum fluid. Then, the governing equations system is simplified in a steady state and the bundle dynamics is simulated, showing that the shape of the velocity profiles depends on two model parameters. Experiments confirm that the model parameters are to be well adjusted to show a coincidence with the theoretical analysis. The higher the drawing ratio and drawing speed we, the more sensitive becomes the bundle flow to exogenous disturbances.

• Communications for Statistical Applications and Methods
• /
• 제19권1호
• /
• pp.1-11
• /
• 2012

로지스틱회귀모형에서 반응변수가 주어졌을 때 설명변수의 조건부 확률분포의 로그-밀도비는 어떤 설명변수가어떻게모형에포함되는지에대한변수선택문제에서유용한정보를제공한다. 설명변수의 조건부 확률분포가 좌우대칭이 아닌 경우 감마분포로 가정하는 것이 적절하다. 여러 가지 모의실험을 수행한 결과를 보면, $x{\mid}y$ = 0과 $x{\mid}y$ = 1의 두 분포가 겹치는 경우에서는 x항과 log(x)항 모두 필요하다. 그리고 두 분포가 분리된 경우에는 x항 또는 log(x)항 중 하나만 필요하다.

• Journal of Animal Science and Technology
• /
• 제59권2호
• /
• pp.3.1-3.7
• /
• 2017

Background: The modelling of lactation curve provides guidelines in formulating farm managerial practices in dairy cows. The aim of the present study was to determine the suitable non-linear model which most accurately fitted to lactation curves of five lactations in 134 Gir crossbred cows reared in Research-CumDevelopment Project (RCDP) on Cattle farm, MPKV (Maharashtra). Four models viz. gamma-type function, quadratic model, mixed log function and Wilmink model were fitted to each lactation separately and then compared on the basis of goodness of fit measures viz. adjusted $R^2$, root mean square error (RMSE), Akaike's Informaion Criteria (AIC) and Bayesian Information Criteria (BIC). Results: In general, highest milk yield was observed in fourth lactation whereas it was lowest in first lactation. Among the models investigated, mixed log function and gamma-type function provided best fit of the lactation curve of first and remaining lactations, respectively. Quadratic model gave least fit to lactation curve in almost all lactations. Peak yield was observed as highest and lowest in fourth and first lactation, respectively. Further, first lactation showed highest persistency but relatively higher time to achieve peak yield than other lactations. Conclusion: Lactation curve modelling using gamma-type function may be helpful to setting the management strategies at farm level, however, modelling must be optimized regularly before implementing them to enhance productivity in Gir crossbred cows.