• Communications for Statistical Applications and Methods
• /
• v.23 no.5
• /
• pp.433-444
• /
• 2016

When we observe binary responses in a cluster (such as rat lab-subjects), they are usually correlated to each other. In clustered binomial counts, the independence assumption is violated and we encounter an extra-variation. In the presence of extra-variation, the ordinary statistical analyses of binomial data are inappropriate to apply. In testing the homogeneity of proportions between several treatment groups, the classical Pearson chi-squared test has a severe flaw in the control of Type I error rates. We focus on modifying the chi-squared statistic by incorporating variance inflation factors. We suggest a method to adjust data in terms of dispersion estimate based on a quasi-likelihood model. We explain the testing procedure via an illustrative example as well as compare the performance of a modified chi-squared test with competitive statistics through a Monte Carlo study.

• Communications for Statistical Applications and Methods
• /
• v.25 no.4
• /
• pp.329-339
• /
• 2018

In this study, we propose tests for equality of several variances with the normality assumption. First of all, we propose the likelihood ratio test by applying the permutation principle. Then by using the p-values for the pairwise tests between variances and combination functions, we propose combination tests. We apply the permutation principle to obtain the overall p-values. Also we review the well- known test statistics for the completion of our discussion and modify a statistic with the p-values. Then we illustrate proposed tests by numerical and simulated data and compare their efficiency with the reviewed ones through a simulation study by obtaining empirical p-values. Finally, we discuss some interesting features related to the resampling methods and tests for equality among several variances.

• Communications of the Korean Mathematical Society
• /
• v.20 no.1
• /
• pp.145-159
• /
• 2005

In this paper we consider the problem of testing statistical hypotheses for unknown parameters in nonlinear regression models and propose three asymptotically equivalent tests based on regression quantiles estimators, which are Wald test, Lagrange Multiplier test and Likelihood Ratio test. We also derive the asymptotic distributions of the three test statistics both under the null hypotheses and under a sequence of local alternatives and verify that the asymptotic relative efficiency of the proposed test statistics with classical test based on least squares depends on the error distributions of the regression models. We give some examples to illustrate that the test based on the regression quantiles estimators performs better than the test based on the least squares estimators of the least absolute deviation estimators when the disturbance has asymmetric and heavy-tailed distribution.

• Journal of the Korean Society of Safety
• /
• v.32 no.6
• /
• pp.46-53
• /
• 2017

The risk is expressed as consequence of damage multiplied by likelihood of failure. The installation of a protective system reduces the risk by reducing the likelihood of failure at the facility. Also, the protective system has different effects on the likelihood of failure according to the proof test cycle. However, when assessing risks in the Off-site Risk Assessment (ORA) system, the variation in risk was not reflected according to the proof test cycle of protective system. This study was conducted to examine the need for proof test and the importance of cycle setting by applying periodic proof test of the protective system to ORA. The results showed that the likelihood of failure and the risk increased with longer proof test cycle. The risk of a two-yearly proof test was eight times greater than that of a three-month cycle. From the results, the protective system needs periodic proof test. Untested protective system for a long term cannot be reliable because it is more likely to be failed state when it is called upon to operate. In order to reduce the risk to an acceptable level, it is effective to differently set the proof test cycle according to the priority. This study suggested a more systematic and accurate risk analysis standard than ORA. This standard is expected to enable an acceptable level of risk management by systematically setting the priority and proof test cycle of the protective system. It is also expected to contribute to securing the safety of chemical facilities and at the same time, will lead to the development of the ORA system.

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.1
• /
• pp.8-14
• /
• 2002

In this paper, we realized a real-time text-independent speaker recognition system using gaussian mixture model, and applied frame level likelihood normalization method which shows its effects in verification system. The system has three parts as front-end, training, recognition. In front-end part, cepstral mean normalization and silence removal method were applied to consider speaker's speaking variations. In training, gaussian mixture model was used for speaker's acoustic feature modeling, and maximum likelihood estimation was used for GMM parameter optimization. In recognition, likelihood score was calculated with speaker models and test data at frame level. As test sentences, we used text-independent sentences. ETRI 445 and KLE 452 database were used for training and test, and cepstrum coefficient and regressive coefficient were used as feature parameters. The experiment results show that the frame-level likelihood method's recognition result is higher than conventional method's, independently the number of registered speakers.

• Journal of Veterinary Clinics
• /
• v.25 no.3
• /
• pp.146-151
• /
• 2008

During the clinical decision making practitioners are often faced with performing diagnostic tests to solve the presenting problems seen in the patients. The diagnostic utility of a test has traditionally been described by technical terms such as sensitivity, specificity, and positive (PPV) and negative predictive value (NPV). Although well known, clinicians are frequently unclear about the concept and application of these terms in everyday evidence-based clinical decision making. Sensitivity and specificity, which are intrinsic properties of diagnostic tests, summarizes the characteristics of the test over a population. The PPV and NPV are greatly dependent on the population prevalence of disease, and thus they do not transferable to different patients or clinical settings. Besides, considering the fact that clinicians more often interested in knowing the extent to which a test result could confirm or exclude of a condition under consideration (posttest probability), these measures do not provide answers on this question. The likelihood ratios (LR) using the information contained in sensitivity and specificity are becoming increasingly popular for reporting the usefulness of diagnostic tests because this term provide an indication of posttest probability as a function of the pretest probability. In this article, clinical applications of LR are illustrated with some practical examples. Discussion is also included of the inherent limitations regarding diagnostic test characteristics.

• Communications for Statistical Applications and Methods
• /
• v.10 no.1
• /
• pp.211-224
• /
• 2003

Good's lambda test, a permutation test used to detect the changes of microorganism composition under two pathological conditions, has been quite popular for studying the micro-flora responsible for periodontal disease. A vast number of different micro-flora in the mouth renders the traditional chi-square test inapplicable. The main purpose of this paper is to evaluate the power of this test so that the sample size can be determined at the design stage. The robustness of this test and its comparison to two other intuitive tests are also presented. It is found that a permutation test based on likelihood ratio is more powerful than the lambda test in our simulated cases.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.8
• /
• pp.47-53
• /
• 2000

In order to classify the underwater vehicles due to propeller propulsion, maximum likelihood classifier was developed. Propeller produces the cavitation and noise during its work. Cavitation-bubble makes the nonlinear medium in the water. The nonlinearity of cavitation leads to the generation of a complete spectrum of combination harmonics of the tonals of noise, and modulation of cavitation noise with propeller shaft-rates and blade-rates. The optimal estimator was derived mathematically and its capabilities were proven by simulation and real test.

• Communications for Statistical Applications and Methods
• /
• v.28 no.3
• /
• pp.307-314
• /
• 2021

We showed that any missing mechanism is reproduced by EMAR or MNAR with equal fit for observed likelihood if there are non-negative solutions of maximum likelihood equations. This is a generalization of Molenberghs et al. (2008) and Jeon et al. (2019). Nonetheless, as MCAR becomes a nested model of MNAR, a natural question is whether or not MNAR and MCAR are testable by using the well-known three statistics, LR (Likelihood ratio), Wald, and Score test statistics. Through simulation studies, we compared these three statistics. We investigated to what extent the boundary solution affect tesing MCAR against MNAR, which is the only testable pair of missing mechanisms based on observed likelihood. We showed that all three statistics are useful as long as the boundary proximity is far from 1.

• Communications for Statistical Applications and Methods
• /
• v.9 no.2
• /
• pp.533-542
• /
• 2002

In this paper, a covariate adjusted logrank test is considered for censored paired data under the Cox proportional hazard model. The proposed score test resembles the adjusted logrank test of Tsiatis, Rosner and Tritchler (1985), which is derived from the partial likelihood. The dependence structure for paired data is accommodated into the test statistic by using' sum of square type' variance estimators. Several weight functions are also considered, which produce a class of covariate adjusted weighted logrank tests. Asymptotic normality of the proposed test is established and simulation studies with moderate sample size show the proposed test works well, particularly when there are dependence structure between treatment and covariates.