• Journal of the Korean Data and Information Science Society
• /
• v.21 no.4
• /
• pp.757-764
• /
• 2010

In this paper, we develop the reference priors for the common location parameter in the half-normal distributions with unequal scale paramters. We derive the reference priors as noninformative prior and prove the propriety of joint posterior distribution under the general prior including the reference priors. Through the simulation study, we show that the proposed reference priors match the target coverage probabilities in a frequentist sense.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.6
• /
• pp.1241-1250
• /
• 2011

In this paper, we develop noninformative priors for the log-normal distributions when the parameter of interest is the common mean. We developed Jeffreys' prior, th reference priors and the first order matching priors. It turns out that the reference prior and Jeffreys' prior do not satisfy a first order matching criterion, and Jeffreys' pri the reference prior and the first order matching prior are different. Some simulation study is performed and a real example is given.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.3
• /
• pp.697-705
• /
• 2003

When X and Y have independent normal distributions with equal coefficient of variation, we develop the reference priors for different groups of ordering for the parameters. Propriety of posteriors under reference priors proved. A real example is presented to compare the classical estimator and Bayes estimator.

• Journal of the Korean Data and Information Science Society
• /
• v.18 no.1
• /
• pp.219-228
• /
• 2007

In this paper, we consider a decision-theoretic oriented, objective Bayesian inference for the ratio of two normal variances. Specifically we derive the Bayesian reference criterion as well as the intrinsic estimator and the credible region which correspond to the intrinsic discrepancy loss and the reference prior. We illustrate our results using real data analysis and simulation study.

• Journal of Veterinary Clinics
• /
• v.27 no.1
• /
• pp.66-70
• /
• 2010

Reference interval is critical for interpreting laboratory results, monitoring response to therapy and predicting the prognosis of the patients in clinical settings. The aim of the present study was to update established reference intervals for routine hematologic and serum chemistry values for a population of clinically healthy dogs (range, 1-8 years) seen in an animal hospital. Blood was obtained by venipuncture while animals were physically restrained, and samples were analyzed for 9 chemistries on MS9-5H (Melot Schloesing Lab, France) and 6 hematology on Vet Test 8008 (IDEXX, USA). Data from 105 dogs (52 males and 53 females) for hematology and 113 dogs (37 males and 76 females) for chemistry were used to determine reference intervals using the parametric, nonparametric and bootstrap methods. Prior to analysis, all parameters were tested for normal distribution using Anderson-Darling criterion. Of the 9 biochemical analytes, alkaline phosphatase, alanine aminotransferase, aspartate aminotransferase, creatinine, total protein, and glucose concentrations did not fit normal distribution for both original and transformed data. All but eosinophil count satisfied normal distribution for either original or transformed data. Parametric method can be used for original cholesterol concentrations, RBC, WBC, and neutrophil counts. This technique can also be used for power-transformed values of blood urea nitrogen concentrations and for logarithm of lymphocyte and monocyte counts. Non-parametric or bootstrap method was the preferred choice for the remaining 7 biochemical parameters and eosinophil count as they did not follow normal distributions. All three statistical techniques performed in similar reference intervals. When establishing reference intervals for clinical laboratory data, it is essential to assess the distribution of the original data to increase the accuracy of the interval, and non-parametric or bootstrap methods are of alternative for the data that do not fit normal distribution.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.4
• /
• pp.1365-1374
• /
• 2006

In this paper, we consider a decision-theoretic oriented, objective Bayesian inference for the difference between two normal means with known variances. We derive the Bayesian reference criterion as well as the intrinsic estimator and the credible region which correspond to the intrinsic discrepancy loss and the reference prior. We show the similarity between derived two-sample results and the results for the one-sample case in Bernardo(1999).

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.2
• /
• pp.297-308
• /
• 2010

This article deals with the problem of testing mean when the coefficient of variation in normal distribution is known. We propose Bayesian hypothesis testing procedures for the normal mean under the noninformative prior. The noninformative prior is usually improper which yields a calibration problem that makes the Bayes factor to be defined up to a multiplicative constant. So we propose the objective Bayesian hypothesis testing procedures based on the fractional Bayes factor and the intrinsic Bayes factor under the reference prior. Specially, we develop intrinsic priors which give asymptotically same Bayes factor with the intrinsic Bayes factor under the reference prior. Simulation study and a real data example are provided.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.4
• /
• pp.1023-1030
• /
• 2003

When X and Y have independent normal distributions, we develop a Bayesian testing procedure for the equality of two coefficients of variation. Under the reference prior of the coefficient of variation, we propose a Bayesian test procedure for the equality of two coefficients of variation using fractional Bayes factor. A real data example is provided.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2004.11a
• /
• pp.272-278
• /
• 2004

Recently, Pepe et al. (2003) employed the receiver operating characteristic (ROC) approach to rank candidate genes from a microarray experiment that can be used for the biomarker development with the ultimate purpose of the population screening of a cancer, In the cancer microarray experiment based on n patients the researcher often wants to compare the tumor tissue with the normal tissue within the same individual using a common reference RNA. This design is referred to as a reference design or an indirect design. Ideally, this experiment produces n pairs of microarray data, where each pair consists of two sets of microarray data resulting from reference versus normal tissue and reference versus tumor tissue hybridizations. However, for certain individuals either normal tissue or tumor tissue is not large enough for the experimenter to extract enough RNA for conducting the microarray experiment, hence there are missing values either in the normal or tumor tissue data. Practically, we have $n_1$ pairs of complete observations, $n_2$ 'normal only' and $n_3$ 'tumor only' data for the microarray experiment with n patients, where n=$n_1$+$n_2$+$n_3$. We refer to this data set as a mixed data set, as it contains a mix of fully observed and partially observed pair data. This mixed data set was actually observed in the microarray experiment based on human tissues, where human tissues were obtained during the surgical operations of cancer patients. Pepe et al. (2003) provide the rationale of using ROC approach based on two independent samples for ranking candidate gene instead of using t or Mann -Whitney statistics. We first modify ROC approach of ranking genes to a paired data set and further extend it to a mixed data set by taking a weighted average of two ROC values obtained by the paired data set and two independent data sets.

• Communications for Statistical Applications and Methods
• /
• v.14 no.1
• /
• pp.11-21
• /
• 2007

In this paper, we consider a decision-theoretic oriented, objective Bayesian inference for the difference between two normal means with unknown com-mon variance. We derive the Bayesian reference criterion as well as the intrinsic estimator and the credible region which correspond to the intrinsic discrepancy loss and the reference prior. We illustrate our results using real data analysis as well as simulation study.