• The Korean Journal of Applied Statistics
• /
• v.20 no.2
• /
• pp.281-290
• /
• 2007

Linear estimator, a weighted sum of the sample observation, is commonly adopted to estimate the finite population parameters such as population totals in survey sampling. The weight for a sampled unit is often constructed by multiplying the base weight, which is the inverse of the first-order inclusion probability, by an adjustment term that takes into account of the auxiliary information obtained throughout the population. The linear estimator using the weight adjustment is often more efficient than the one using only the bare weight, but its valiance estimation is more complicated. We discuss variance estimation for a general class of weight-adjusted estimator. By identifying that the weight-adjusted estimator can be viewed as a function of estimated nuisance parameters, where the nuisance parameters were used to incorporate the auxiliary information, we derive a linearization of the weight-adjusted estimator using a Taylor expansion. The method proposed here is quite general and can be applied to wide class of the weight-adjusted estimators. Some examples and results from a simulation study are presented.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.5
• /
• pp.1153-1165
• /
• 2017

Structural equation modeling (SEM) is a basic tool for testing theories in a variety of disciplines. A maximum likelihood (ML) method for parameter estimation is by far the most widely used in SEM. Alternatively, two-stage least squares (2SLS) estimator has been proposed as a more robust procedure to address model misspecification. A regularized extension of 2SLS, two-stage ridge least squares (2SRLS) has recently been introduced as an alternative to ML to effectively handle the small-sample-size issue. However, it is unclear whether and when misspecification and small sample sizes may pose problems in theory testing with 2SLS, 2SRLS, and ML. The purpose of this article is to evaluate the three estimation methods in terms of inferences errors as well as parameter recovery under two experimental conditions. We find that: 1) when the model is misspecified, 2SRLS tends to recover parameters better than the other two estimation methods; 2) Regardless of specification errors, 2SRLS produces small or relatively acceptable Type II error rates for the small sample sizes.

• Journal of the Korean Data and Information Science Society
• /
• v.8 no.2
• /
• pp.153-161
• /
• 1997

A Korean exporience life table(male) is constructed by using a mixture of weighted moving average(WMA) model and Gompertz' parametric survival model based on 25,000,000 insured of major 6 life insurance companies from 1988 to 1992. The graduated values are taken as those which minimize the composite measure of fittness and smoothness. Moreover, we propose closed form estimators for three parameters of Gompertz' model.

• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.665-679
• /
• 2017

When measuring productive efficiency, often it is necessary to have knowledge of the production frontier function that shows the maximum possible output of production units as a function of inputs. Canonical parametric forms of the frontier function were initially considered under the framework of stochastic frontier model; however, several additional nonparametric methods have been developed over the last decade. Efforts have been recently made to impose shape constraints such as monotonicity and concavity on the non-parametric estimation of the frontier function; however, most existing methods along that direction suffer from unnecessary non-smooth points of the frontier function. In this paper, we propose methods to estimate the smooth frontier function with monotonicity for stochastic frontier models and investigate the effect of imposing a monotonicity constraint into the estimation of the frontier function and the finite dimensional parameters of the model. Simulation studies suggest that imposing the constraint provide better performance to estimate the frontier function, especially when the sample size is small or moderate. However, no apparent gain was observed concerning the estimation of the parameters of the error distribution regardless of sample size.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.2
• /
• pp.377-385
• /
• 2015

After performing callback for nonresponses in sample survey, we present an estimator of regression form using an auxiliary variable and a variance estimator using replicate method. Parametric inference method of the response probability is also presented. We research an unbiased estimator of high efficiency for the population mean and a variance estimator with consistency under callback. We also prove the validity of the theory through the simulation.

• The Korean Journal of Applied Statistics
• /
• v.29 no.5
• /
• pp.807-822
• /
• 2016

The composite lognormal-GPD models (LN-GPD) enjoys both merits from log-normality for the body of distribution and GPD for the thick tailedness of the observation. However, in the estimation perspective, LN-GPD model performs poorly due to numerical instability. Therefore, a two-stage procedure, that estimates threshold first then estimates other parameters later, is a natural method to consider. This paper considers five nonparametric threshold estimation methods widely used in extreme value theory and compares their performance in LN-GPD parameter estimation. A simulation study reveals that simultaneous maximum likelihood estimation performs good in threshold estimation, but very poor in tail index estimation. However, the nonparametric method performs good in tail index estimation, but introduced bias in threshold estimation. Our method is illustrated to the service time of an Israel bank call center and shows that the LN-GPD model fits better than LN or GPD model alone.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.21 no.6
• /
• pp.480-494
• /
• 2009

Long term extreme wave data are essential for planning and designing coastal structures. Since the availability of the field data for the waters around Korean peninsula is limited to provide a reliable wave statistics, the wave climate information has been generated by means of long-term wave hindcasting using available meteorological data. KORDI(2005) has proposed extreme wave data at 106 stations off the Korean coast from 1979 to 2003. In this paper, extreme data sets of wave(KORDI, 2005) have been analyzed for best-fitting distribution functions, for which the spread parameter proposed by Goda(2004) is evaluated. The calculated values of the spread parameter are in good agreement with the values based on method of moment for parameter estimation. However, the spread parameter of extreme wave data has a representative value ranging from about 1.0 to 2.8 which is larger than some foreign coastal waters, it is necessary to review deep water design wave.

• Journal of the Korean Data and Information Science Society
• /
• v.11 no.2
• /
• pp.235-245
• /
• 2000

In this paper, we consider the Regression Quantiles Estimators in nonlinear regression models. This paper provides the sufficient conditions for strong consistency and asymptotic normality of proposed estimation and drives asymptotic relative efficiency of proposed estimatiors with least square estimation. We give some examples and results of Monte Carlo simulation to compare least square and regression quantile estimators.

• Convergence Security Journal
• /
• v.12 no.1
• /
• pp.9-16
• /
• 2012

Due to the large scale application of software systems, software reliability plays an important role in software developments. In this paper, a software reliability growth model (SRGM) is proposed for testing time. The testing time on the right is truncated in this model. The intensity function, mean-value function, reliability of the software, estimation of parameters and the special applications of Pareto NHPP model are discussed. This paper, a numerical example of applying using time between failures and parameter estimation using maximum likelihood estimation method, after the efficiency of the data through trend analysis model selection, depended on difference between predictions and actual values, were efficient using the mean square error and $R_{SQ}$.

• Journal of Korea Technology Innovation Society
• /
• v.6 no.3
• /
• pp.279-297
• /
• 2003

The purpose of this paper is to analyze the determinants of company R＆D investment with zero observations by using the data of R＆D Scoreboard published by Ministry of Science and Technology(2002). Conventional parametric approach to dealing with zero investments is not robust to heteroscedastic and/or non-normal error structure. Thus, this study applies symmetrically trimmed least squares(STLS) estimation as a semi-parametric approach to dealing with zero R&D investments. The result of specification test indicates the semi-parametric approach outperforms the parametric approach significantly. Moreover, the results of the study provide various implications as summarized below. The R&D investment of IT company is larger than that of non-IT company. The R&D investment has a positive relation to foreigners' investment ratio. The higher degree of financial self-reliance is, the larger the R&D investment is. Firm size variables such as sales amount and the number of workers are positively related to R&D investment. The sales elasticity of R&D investment is larger than one. However, the workers elasticity of R&D investment is smaller than one.