• Journal of Korea Water Resources Association
• /
• v.42 no.4
• /
• pp.331-341
• /
• 2009

In this study, we compared the observed information matrix with the Fisher information matrix to estimate the uncertainty of maximum likelihood estimators of the generalized logistic (GL) distribution. The previous literatures recommended the use of the observed information matrix because this is convenient since this matrix is determined as the part of the parameter estimation procedure and there is little difference in accuracy between the observed information matrix and the Fisher information matrix for large sample size. The observed information matrix has been applied for the generalized logistic distribution based on the previous study without verification. For this purpose, a simulation experiment was performed to verify which matrix gave the better accuracy for the GL model. The simulation results showed that the variance-covariance of the ML parameters for the GL distribution came up with similar results to those of previous literature, but it is preferable to use of the Fisher information matrix to estimate the uncertainty of quantile of ML estimators.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.3
• /
• pp.522-538
• /
• 2013

Content based image retrieval has become an increasingly important research topic for its wide application. It is highly challenging when facing to large-scale database with large variance. The retrieval systems rely on a key component, the predefined or learned similarity measures over images. We note that, the similarity measures can be potential improved if the data distribution information is exploited using a more sophisticated way. In this paper, we propose a similarity measure learning approach for image retrieval. The similarity measure, so called Fisher kernel, is derived from the probabilistic distribution of images and is the function over observed data, hidden variable and model parameters, where the hidden variables encode high level information which are powerful in discrimination and are failed to be exploited in previous methods. We further propose a discriminative learning method for the similarity measure, i.e., encouraging the learned similarity to take a large value for a pair of images with the same label and to take a small value for a pair of images with distinct labels. The learned similarity measure, fully exploiting the data distribution, is well adapted to dataset and would improve the retrieval system. We evaluate the proposed method on Corel-1000, Corel5k, Caltech101 and MIRFlickr 25,000 databases. The results show the competitive performance of the proposed method.

• Journal of the Korean Statistical Society
• /
• v.17 no.2
• /
• pp.93-100
• /
• 1988

Two properties of ancillary statistics are considered. One is to find a role of ancillary statistics in the statistical inference by showing that the ancillary statistic can recover the lost information and to give a criteria for comparing the conditional inference with unconditional inference. The other is to find an ancillary statistic of translation model and its relationship with observed Fisher information.

• Communications for Statistical Applications and Methods
• /
• v.25 no.4
• /
• pp.411-430
• /
• 2018

The Bayesian approach is a suitable alternative in constructing appropriate models for observed record values because the number of these values is small. This paper provides an objective Bayesian analysis method for upper record values arising from the Rayleigh distribution. For the objective Bayesian analysis, the Fisher information matrix for unknown parameters is derived in terms of the second derivative of the log-likelihood function by using Leibniz's rule; subsequently, objective priors are provided, resulting in proper posterior distributions. We examine if these priors are the PMPs. In a simulation study, inference results under the provided priors are compared through Monte Carlo simulations. Through real data analysis, we reveal a limitation of the appropriate confidence interval based on the maximum likelihood estimator for the scale parameter and evaluate the models under the provided priors.

• Asian Pacific Journal of Cancer Prevention
• /
• v.13 no.6
• /
• pp.2931-2934
• /
• 2012

Background: The epidemiological situation regarding cervical carcinoma in Serbia is rather unfavorable and one of contributing factors is the insufficient interest of women concerning the risk factors responsible for occurrence of this disease. The aim of this study was to determine the sources of relevant information for women Methods: An anonymous questionnaire was used for questioning of patients, students and women undergoing systematic examinations. There were 600 women in total in 2006, 2009 and 2010, and the data were statistically processed by the ${\chi}^2$ test with Yates correction and the Fisher test. Results: When observed for certain groups of tested women, and summed up for all three periods, there was a statistically significant difference for the answer "without any knowledge" (p=0.0001). When observed for certain years and summed up for all three tested groups, there was a statistically significant difference in answers regarding the source of information, the "doctor" (p=0.0011), "media" (p=0.0349) and "encyclopedia-internet" (p=0.0136). Conclusion: The media are a dominant source of information for women on risk factors for cervical cancer. The significance of the Internet increased during the three observed periods, while the students considered themselves least informed of all concerning risk factors.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.5
• /
• pp.1039-1055
• /
• 2014

The distribution of a test statistic under a null hypothesis depends on the unknown distribution of the data and thus is unknown as well. Conditional tests replace the unknown null distribution by the conditional null distribution, that is, the distribution of the test statistic given the observed data. This approach is known as permutation tests and was developed by Fisher (Fisher, 1935). Theoretical framework for permutation tests was given by Strasser and Weber(1999). The coin package developed by Hothon et al. (2006, 2008) implements a unified approach for conditional inference via the generic independence test. Because convenient functions for the most prominent problems are available, users will not have to use the extremely flexible procedure. In this article we briefly review the underlying theory from Strasser and Weber (1999) and explain how to transform the data to perform the generic function independence test. Finally it was illustrated with a few real data sets.

• Communications for Statistical Applications and Methods
• /
• v.24 no.3
• /
• pp.291-301
• /
• 2017

It is well known in a small sample that the maximum likelihood (ML) approach for variance components in the general linear model yields estimates that are biased downward. The ML estimate of residual variance tends to be downwardly biased. The underestimation of residual variance, which has implications for the estimation of marginal effects and asymptotic standard error of estimates, seems to be more serious in some limited dependent variable models, as shown by some researchers. An alternative frequentist's approach may be restricted or residual maximum likelihood (REML), which accounts for the loss in degrees of freedom and gives an unbiased estimate of residual variance. In this situation, the REML estimator is derived in a censored regression model. A small sample the REML is shown to provide proper inference on regression coefficients.

• Communications for Statistical Applications and Methods
• /
• v.25 no.6
• /
• pp.647-658
• /
• 2018

In this paper, we study statistical inferences on the maximum likelihood estimation of a normal distribution when data are randomly censored. Likelihood equations are derived assuming that the censoring distribution does not involve any parameters of interest. The maximum likelihood estimators (MLEs) of the censored normal distribution do not have an explicit form, and it should be solved in an iterative way. We consider a simple method to derive an explicit form of the approximate MLEs with no iterations by expanding the nonlinear parts of the likelihood equations in Taylor series around some suitable points. The points are closely related to Kaplan-Meier estimators. By using the same method, the observed Fisher information is also approximated to obtain asymptotic variances of the estimators. An illustrative example is presented, and a simulation study is conducted to compare the performances of the estimators. In addition to their explicit form, the approximate MLEs are as efficient as the MLEs in terms of variances.

• Journal of dental hygiene science
• /
• v.18 no.2
• /
• pp.76-84
• /
• 2018

Wet wipes are being increasingly used because of their convenience. Particularly, oral wet wipes are useful for regular cleaning of a baby's mouth after birth. Therefore, the consumption of oral wet wipes has increased over the past few years and a variety of products are commercially available. However, product information on safety is not sufficiently provided and still raises doubts regarding adverse effects. To confirm the safety of wet wipes as an oral hygiene item and provide information for their use, we investigated the cytotoxicity of oral wet wipes and verified the underlying mechanism. The anti-bacterial effect of oral wet wipes was analyzed using the disk diffusion method. The cytotoxic effects of oral wet wipes were observed based on morphological changes using microscopy and determined using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay in gingival epithelial cells and gingival fibroblasts. Evaluation of apoptosis by oral wet wipes was explored using propidium iodide flow cytometric analysis and a terminal deoxynucleotidyl transferase (TdT)-mediated deoxyuridine triphosphate (dUTP) nick-end labeling (TUNEL) assay. Apoptosis-related molecules were also analyzed using western blotting. Five types of oral wet wipes were tested, and two products from Fisher-Price and Dr. Kennedy revealed strong cytotoxic effects on gingiva epithelial cells and gingiva fibroblasts, although they also showed intense anti-bacterial effects on oral bacteria. Cell cycle arrest in the G2/M phase and apoptosis were observed based on treatment of extracts from Fisher-Price and Dr. KENNEDY. Relatively high TUNEL levels, reduction of proliferating cell nuclear antigen and cyclin-dependent kinase 4 expression, and fragmentation of poly (ADP-ribose) polymerase were also elucidated. These results suggest that commercial oral wet wipes could exert cytotoxic influences on oral tissue, although there are anti-bacterial effects, and careful attention is required, especially for infants and toddlers.

• Korean Journal of Clinical Pharmacy
• /
• v.27 no.3
• /
• pp.171-177
• /
• 2017

Objective: The study purpose was to develop a drug information leaflet for the elderly and to evaluate it with performance-based user-testing. Methods: We performed a stratified randomized controlled trial. We recruited 62 elderly patients with age of 65 or above who were taking antidiabetic medications at the point of participating and excluded those who suffered illiteracy. We randomly allocated them into the intervention group with a leaflet for the elderly and the control group with a leaflet for the general public. Main outcome measures were to 'be able to find information' and to 'be able to understand information.' We measured outcome variables by employing performance-based user-testing and analyzed data to find any differences between two groups with t-tests, chi-squared tests or Fisher's exact tests accordingly. Results: More participants in the intervention group understood how to store their medications than those in the control group (intervention group 93% vs. control group 70%; p=0.02). There were no significant differences in other information items between two groups. Mostly 'being able to understand information' was lower than 'being able to find information.' The gaps between two outcome variables were about 10% in the intervention group and about 18% in the control group. The lowest understanding was observed in information relating to drug names and their potential adverse events. Conclusion: Without providing personalized drug information, it might be hard for the elderly to improve their drug knowledge even with leaflets that were developed specifically for the elderly.