• Journal of the Korean Statistical Society
• /
• v.33 no.4
• /
• pp.411-433
• /
• 2004

The shrinkage preliminary test ridge regression estimators (SPTRRE) based on Wald (W), Likelihood Ratio (LR) and Lagrangian Multiplier (LM) tests for estimating the regression parameters of the multiple linear regression model with multivariate Student's t error distribution are considered in this paper. The quadratic biases and risks of the proposed estimators are compared under both null and alternative hypotheses. It is observed that there is conflict among the three estimators with respect to their risks because of certain inequalities that exist among the test statistics. In the neighborhood of the restriction, the SPTRRE based on LM test has the smallest risk followed by the estimators based on LR and W tests. However, the SPTRRE based on W test performs the best followed by the LR and LM based estimators when the parameters move away from the subspace of the restrictions. Some tables for the maximum and minimum guaranteed efficiency of the proposed estimators have been given, which allow us to determine the optimum level of significance corresponding to the optimum estimator among proposed estimators. It is evident that in the choice of the smallest significance level to yield the best estimator the SPTRRE based on Wald test dominates the other two estimators.

• Communications for Statistical Applications and Methods
• /
• v.11 no.3
• /
• pp.475-484
• /
• 2004

This paper consider the testing problem of grouped heteroscedasticity in the linear regression model. We provide the Lagrange Multiplier(LM), Wald, Likelihood Ratio (LR) test statistis for testing of grouped heteroscedasticity. Monte Carlo experiments are conducted to study the performance of these tests.

• Communications for Statistical Applications and Methods
• /
• v.15 no.2
• /
• pp.161-171
• /
• 2008

This study discusses Johnson's $S_U$-normal distribution capturing a wide range of non-normality in various regression models. We provide the likelihood inference using Johnson's $S_U$-normal distribution, and propose a likelihood ratio (LR) test for normality. We also apply the $S_U$-normal distribution to the binary and censored regression models. Monte Carlo simulations are used to show that the LR test using the $S_U$-normal distribution can be served as a model specification test for normal error distribution, and that the $S_U$-normal maximum likelihood (ML) estimators tend to yield more reliable marginal effect estimates in the binary and censored model when the error distributions are non-normal.

• Journal of Korean Academy of Nursing
• /
• v.43 no.2
• /
• pp.154-164
• /
• 2013

Purpose: The purpose of this article is twofold: 1) introducing logistic regression (LR), a multivariable method for modeling the relationship between multiple independent variables and a categorical dependent variable, and 2) examining use and reporting of LR in the nursing literature. Methods: Text books on LR and research articles employing LR as main statistical analysis were reviewed. Twenty-three articles published between 2010 and 2011 in the Journal of Korean Academy of Nursing were analyzed for proper use and reporting of LR models. Results: Logistic regression from basic concepts such as odds, odds ratio, logit transformation and logistic curve, assumption, fitting, reporting and interpreting to cautions were presented. Substantial shortcomings were found in both use of LR and reporting of results. For many studies, sample size was not sufficiently large to call into question the accuracy of the regression model. Additionally, only one study reported validation analysis. Conclusion: Nursing researchers need to pay greater attention to guidelines concerning the use and reporting of LR models.

• Journal of Korean Society for Quality Management
• /
• v.23 no.4
• /
• pp.52-57
• /
• 1995

Testing problem of equality between two linear regressions is considered. Chow test is discussed and likelihood ratio test(LR test), F-test and Score test are derived. Score test are compared with other tests and show they are asymptotically equivalent.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.6
• /
• pp.111-117
• /
• 2007

This paper presents a novel approach to speech enhancement based on a complex Laplacian probability density function (pdf). With a use of goodness-of-fit (GOF) test we show that the complex Laplacian pdf is more suitable to describe the conventional Gaussian pdf. The likelihood ratio (LR) is applied to derive the speech absence probability in the speech enhancement algorithm. The performance of the proposed algorithm is evaluated by the objective test and yields better results compared with the conventional Gaussian pdf-based scheme.

• 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.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.1
• /
• pp.97-103
• /
• 2012

In this paper, we propose a novel voice activity detection (VAD) algorithm to effectively distinguish speech from nonspeech in various noisy environments. Global speech absence probability (GSAP) derived from likelihood ratio (LR) based on the statistical model is widely used as the feature parameter for VAD. However, the feature parameter based on conventional GSAP is not sufficient to distinguish speech from noise at low SNRs (signal-to-noise ratios). The presented VAD algorithm utilizes GSAP based on Teager energy (TE) as the feature parameter to provide the improved performance of decision for speech segments in noisy environment. Performances of the proposed VAD algorithm are evaluated by objective test under various environments and better results compared with the conventional methods are obtained.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.16
• /
• pp.6571-6574
• /
• 2014

Background: Alcohol is well established as a risk factor for cancer development in many organ sites. To assess the reliability and validity of the Alcohol Use Disorders Identification Test - Consumption (AUDIT-C) for detecting alcohol use disorders or risky drinking in Japanese adults the present study was conducted. Materials and Methods: A test-retest method was applied with a 2-week interval with 113 health care employees. The k coefficient, Cronbach's coefficient alpha, Spearman's correlation coefficient, and intraclass correlation coefficient (ICC) were determined and the validity of the AUDIT-C was analyzed using the data from a nationwide survey on adult alcohol use conducted in 2008 (n=4,123). Results: The reliability of the AUDIT-C score was high (${\kappa}$ coefficient=0.63, Cronbach's alpha=0.98, correlation coefficient=0.95, and ICC=0.95). According to the likelihood ratio and Youden index, appropriate cutoffs for the AUDIT-C were ${\geq}5points$ in men and ${\geq}4$ points in women. The sensitivity and specificity of these cutoffs for identifying ${\geq}8$ points on the AUDIT were 0.88 and 0.80, respectively, for men (positive likelihood ratio [LR+]=4.5) and 0.96 and 0.87, respectively, for women (LR+=7.7). The sensitivity and specificity of the cutoffs for identifying ${\geq}12$ points on the AUDIT were 0.90 and 0.84, respectively, for men (LR+=5.8) and 0.93 and 0.94, respectively, for women (LR+=15.8). The sensitivity and specificity of the cutoffs for identifying ${\geq}16$ points on the AUDIT were 0.93 and 0.80, respectively, for men (LR+=4.7) and 0.92 and 0.98, respectively, for women (LR+=55.6). With higher scores on the AUDIT, the specificity decreased and false-positives increased. The appropriate cutoffs for identifying risky drinking were the same for both genders. Conclusions: The reliability and validity of the AUDIT-C are high, indicating that it is useful for identifying alcohol use disorders or risky drinking among the general population in Japan, a group at high risk of cancer development.

• 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.