• Journal of the Korean Statistical Society
• /
• v.24 no.1
• /
• pp.161-181
• /
• 1995

The goal of this article is to study the performances of various empirical Bayes simultaneous interval estimates for all pairwise comparisons. The considered empirical Bayes interval estimaters are those based on unbiased estimate, a hierarchical Bayes estimate and a constrained hierarchical Bayes estimate. Simulation results for small sample cases are given and an illustrative example is also provided.

• Communications for Statistical Applications and Methods
• /
• v.20 no.4
• /
• pp.321-327
• /
• 2013

Bayesian and empirical Bayesian methods have become quite popular in the theory and practice of statistics. However, the objective is to often produce an ensemble of parameter estimates as well as to produce the histogram of the estimates. For example, in insurance pricing, the accurate point estimates of risk for each group is necessary and also proper dispersion estimation should be considered. Well-known Bayes estimates (which is the posterior means under quadratic loss) are underdispersed as an estimate of the histogram of parameters. The adjustment of Bayes estimates to correct this problem is known as constrained Bayes estimators, which are matching the first two empirical moments. In this paper, we propose a way to apply the constrained Bayes estimators in insurance pricing, which is required to estimate accurately both location and dispersion. Also, the benefit of the constrained Bayes estimates will be discussed by analyzing real insurance accident data.

• The Korean Journal of Physiology and Pharmacology
• /
• v.16 no.2
• /
• pp.97-106
• /
• 2012

The pharmacokinetics/pharmacodynamics analysis software NONMEM$^{(R)}$ output provides model parameter estimates and associated standard errors. However, the standard error of empirical Bayes estimates of inter-subject variability is not available. A simple and direct method for estimating standard error of the empirical Bayes estimates of inter-subject variability using the NONMEM$^{(R)}$ VI internal matrix POSTV is developed and applied to several pharmacokinetic models using intensively or sparsely sampled data for demonstration and to evaluate performance. The computed standard error is in general similar to the results from other post-processing methods and the degree of difference, if any, depends on the employed estimation options.

• Communications for Statistical Applications and Methods
• /
• v.21 no.3
• /
• pp.235-243
• /
• 2014

Constrained Bayesian estimates overcome the over shrinkness toward the mean which usual Bayes and empirical Bayes estimates produce by matching first and second empirical moments; subsequently, a constrained Bayes estimate is recommended to use in case the research objective is to produce a histogram of the estimates considering the location and dispersion. The well-known squared error loss function exclusively emphasizes the precision of estimation and may lead to biased estimators. Thus, the balanced loss function is suggested to reflect both goodness of fit and precision of estimation. In insurance pricing, the accurate location estimates of risk and also dispersion estimates of each risk group should be considered under proper loss function. In this paper, by applying these two ideas, the benefit of the constrained Bayes estimates and balanced loss function will be discussed; in addition, application effectiveness will be proved through an analysis of real insurance accident data.

• The Korean Journal of Applied Statistics
• /
• v.6 no.1
• /
• pp.173-181
• /
• 1993

Nonparametric empirical Bayes estimation of a distribution function with respect to dirichlet process prior is considered when sample sizes are varying from component to component. Zehnwirth's estimate of $\alpha$(R) is modified to be used in our empirical Bayes problem with non-identical components.

• Communications for Statistical Applications and Methods
• /
• v.9 no.2
• /
• pp.515-520
• /
• 2002

In general, we use the hierarchical Poisson-gamma model for the Poisson data in generalized linear model. Time effect will be emphasized for the analysis of the observed data to be collected annually for the time period. An extended model with time effect for estimating the effect is proposed. In particularly, we discuss the Quasi likelihood function which is used to numerical approximation for the likelihood function of the parameter.

• Communications for Statistical Applications and Methods
• /
• v.9 no.2
• /
• pp.381-388
• /
• 2002

It has been pointed out that the classical credibility model used in Korea since the beginning of 1990's lacks in objectiveness. Recently, in order to improve objectiveness, the empirical Bayes credibility model utilizing general exposure units like the number of claims and premium has been employed, but that model itself is not quite applicable in the country like Korea whose annual and classified empirical data are not well accumulated and even varied severely. In this article, we propose a new and better model, Based on the new model, we estimate both credibility and loss ratio of each class for fire insurance plans by Korean insurance companies. As a conclusion, we empirically make sure analysis that the number of claims is a more reasonable exposure unit than premium.

• Journal of the Korean Statistical Society
• /
• v.25 no.4
• /
• pp.529-544
• /
• 1996

This paper deals with the problem of estimating the autoregressive random coefficient of a first-order random coefficient autoregressive time series model applied to panel data of time series. The autoregressive random coefficients across individual units are assumed to be a random sample from a truncated normal distribution with the space (-1, 1) for stationarity. The estimates of random coefficients are obtained by an empirical Bayes procedure using the estimates of model parameters. Also, a Monte Carlo study is conducted to support the estimation procedure proposed in this paper. Finally, we apply our results to the economic panel data in Liu and Tiao(1980).

• Communications for Statistical Applications and Methods
• /
• v.6 no.1
• /
• pp.221-235
• /
• 1999

Let be $X_1$,...,$X_p$, $p\geq2$ independent random variables where each $X_i$ has a gamma distribution with $\textit{k}_i$ and $\theta_i$ The problem is to simultaneously estimate $\textit{p}$ gamma parameters $\theta_i$ and $\theta_i{^-1}$ under entropy loss where the parameters are believed priori. Hierarch ical Bayes(HB) and empirical Bayes(EB) estimators are investigated. And a preference of HB estimator over EB estimator is shown using Gibbs sampler(Gelfand and Smith 1990). Finally computer simulation is studied to compute the risk percentage improvements of the HB estimator and the estimator of Dey Ghosh and Srinivasan(1987) compared to UMVUE estimator of $\theta^{-1}$.

• Journal of applied mathematics & informatics
• /
• v.3 no.1
• /
• pp.55-64
• /
• 1996

Let $X_1, ....$X_P be p($\geq$2) independent random variables, where each X1 has a gamma distribution with $k_i and ${\heta}_i$. The problem is to simultaneously estimate p gammar parameters ${\heta}_i$ under entropy loss where the parameters are believed priori. Hierarchical bayes(HB) and empirical bayes(EB) estimators are investigated. Next computer simulation is studied to compute the risk percentage improvement of the HB, EB and the estimator of Dey et al.(1987) compared to MVUE of ${\heta}$.