• The Korean Journal of Applied Statistics
• /
• v.30 no.6
• /
• pp.957-981
• /
• 2017

A generalized linear mixed model is an extension of a generalized linear model that allows random effect as well as provides flexibility in developing a suitable model when observations are correlated or when there are other underlying phenomena that contribute to resulting variability. We describe maximum likelihood estimation methods for logistic regression models that include random effects - the Laplace approximation, Gauss-Hermite quadrature, adaptive Gauss-Hermite quadrature, and pseudo-likelihood. Applications are provided with social science problems by analyzing the effect of mental health and life satisfaction on volunteer activities from Korean welfare panel data; in addition, we observe that the inclusion of random effects in the model leads to improved analyses with more reasonable inferences.

• Journal of KIISE:Software and Applications
• /
• v.36 no.4
• /
• pp.273-282
• /
• 2009

Missing data is one of the common problems in building analysis or prediction models using software project data. Missing imputation methods are known to be more effective missing data handling method than deleting methods in small software project data. While K nearest neighbor imputation is a proper missing imputation method in the software project data, it cannot use non-missing information of incomplete project instances. In this paper, we propose an approach to missing data imputation for numerical software project data by combining K nearest neighbor and maximum likelihood estimation; we also extend the average absolute error measure by normalization for accurate evaluation. Our approach overcomes the limitation of K nearest neighbor imputation and outperforms on our real data sets.

• Communications for Statistical Applications and Methods
• /
• v.17 no.1
• /
• pp.67-77
• /
• 2010

In multinomial scheme with step-wise sampling, maximum likelihood estimates of multinomial probabilities are improved when some frequencies are merged. In this study, for cell probabilities in a I by J independent contingency tables, exact MLE and step-wise estimation methods are applied and the results are compared using MSE and Bias.

• Journal of the Korean Data and Information Science Society
• /
• v.11 no.2
• /
• pp.295-302
• /
• 2000

We study the efficient method to estimate the parameters for the Weibull model with covariates which occupies an important position in survival analysis. A treatment period may be divided by the stages of treatments under the different treatment arams. The piecewise method is considered to obtain the estimators of the parameters by maximum likelihood method. We explore the real data to show that the piecewise is more efficient than the nonpiecewise to estimate the parameters.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.6
• /
• pp.1309-1317
• /
• 2013

There are various parameter estimation methods for the generalized exponential distribution under progressive type I interval censoring. Chen and Lio (2010) studied the parameter estimation method by the maximum likelihood estimation method, mid-point approximation method, expectation maximization algorithm and methods of moments. Among those, mid-point approximation method has the smallest mean square error in the generalized exponential distribution under progressive type I interval censoring. However, this method is difficult to derive closed form of solution for the parameter estimation using by maximum likelihood estimation method. In this paper, we propose two type of approximate maximum likelihood estimate to solve that problem. The simulation results show the obtained estimators have good performance in the sense of the mean square error. And proposed method derive closed form of solution for the parameter estimation from the generalized exponential distribution under progressive type I interval censoring.

• The Korean Journal of Applied Statistics
• /
• v.28 no.5
• /
• pp.895-906
• /
• 2015

Diffusion is a mathematical tool to explain the fluctuation of financial assets and the movement of particles in a micro time scale. There are ongoing statistical trials to develop an estimation method for diffusion models based on likelihood. When we estimate diffusion models by applying the maximum likelihood estimation method on data observed at discrete time points, we need to know the transition density of the diffusion. In order to approximate the transition densities of diffusion models, we suggests the method to approximate the path integral of the random process with normal random variables, and compare the numerical properties of the method with other approximation methods.

• The Journal of Engineering Geology
• /
• v.3 no.1
• /
• pp.51-61
• /
• 1993

Restricted maximum likelihood(RML) method was used to determine the parameters of generalized covariance, and universal krigig with RML was applied to estimate a groundwater level distribution of nonstationarv random function. Universal kriging with RML was compared to IRF-k with weighted least squares method for the comparison of their accuracies. Cross validation shows that two methods have nearly the same ability for the estimation of groundwater levels. Scattergram of estimates versus true values and contour maps of groundwater levels have nearly the same results. The reason why two methods produced the same results is thought to be the non-Gaussian distribution and the snaall number of sample data.

• Journal of the Korean Data and Information Science Society
• /
• v.7 no.2
• /
• pp.263-272
• /
• 1996

In this parer, under random censorship model, an estimation of scale and shape parameters in Weibull lifetime model is considered. Based on nonparametric estimator of survival function, the least square method is proposed. The proposed estimation method is simple and the performance of the proposed estimator is as efficient as maximum likelihood estimators. An example is presented, using field winding data. Simulation studies are performed to compare the performaces of the proposed estimator and maximum likelihood estimator.

• Communications for Statistical Applications and Methods
• /
• v.16 no.4
• /
• pp.723-730
• /
• 2009

Missing values in time series can be treated as unknown parameters and estimated by maximum likelihood or as random variables and predicted by the expectation of the unknown values given the data. The purpose of this study is to impute missing values which are regarded as the maximum likelihood estimator and random variable in incomplete data and to compare with two methods using ARMA model. For illustration, the Mumps data reported from the national capital region monthly over the years 2001 ${\sim}$ 2006 are used, and results from two methods are compared with using SSF(Sum of square for forecasting error).

• The Korean Journal of Applied Statistics
• /
• v.26 no.3
• /
• pp.431-440
• /
• 2013

We consider the problem of estimating the parameters of heavy tailed distributions. In general, maximum likelihood estimation(MLE) is the most preferred method of parameter estimation because it has good properties such as asymptotic consistency, normality and efficiency. However, MLE is not always the best solution because MLE is unstable or does not exist in some cases. This paper proposes another parameter estimation method, non-linear least squares(NLS) and compares its performance to MLE. The NLS estimator is achieved by minimizing sum of squared difference between empirical cumulative distribution function(CDF) and a theoretical distribution function. In this article, we compare the NLS method to MLE using simulated data from heavy tailed distributions. The NLS method is shown to perform better than MLE in Burr distribution when the sample size is small; in addition, it performs well in a Frechet distribution.