• 응용통계연구
• /
• 제31권4호
• /
• pp.475-484
• /
• 2018

감마 일반화 선형모형은 음이 아니며 치우침이 있는 반응변수에 유용한 모형으로 알려져 있다. 그러나 포아송 분포 또는 이항 분포에 기반한 일반화 선형모형에 비해 적은 관심을 받아왔다. 특히, 회귀계수의 유의성 검정에 대해서는 연구가 면밀히 되어 있지 않다. 본 논문에서는 감마 일반화 선형 모형의 검정에 대해 다양한 통계량들을 알아보고 수치 연구를 통해 그들의 성능을 비교한다. 수치 실험의 결과 부분 이탈도 검정 방법의 문제점이 나타났으며, 가능도비 검정 방법과 F-검정 방법이 좋은 성능을 보임을 확인하였다.

• Journal of Electrical Engineering and Technology
• /
• 제6권4호
• /
• pp.563-572
• /
• 2011

Reliability is an essential factor in a navigation system. Therefore, an integrity monitoring system is considered one of the most important parts in an avionic navigation system. A fault due to systematic malfunctioning definitely requires integrity reinforcement through systematic analysis. In this paper, we propose a method to detect faults of the GPS signal by using a distributed nonlinear filter based probability test. In order to detect faults, consistency is examined through a likelihood ratio between the main and auxiliary particle filters (PFs). Specifically, the main PF which includes all the measurements and the auxiliary PFs which only do partial measurements are used in the process of consistency testing. Through GPS measurement and the application of the autonomous integrity monitoring system, the current study illustrates the performance of the proposed fault detection algorithm.

• Journal of the Korean Statistical Society
• /
• 제24권2호
• /
• pp.419-437
• /
• 1995

Given the specific mean shift outlier model, several standard approaches to obtaining test statistic for outliers are discussed. Each of these is developed in detail for the nonlinear regression model, and each leads to an equivalent distribution. The geometric interpretations of the statistics and accuracy of linear approximation are also presented.

• Communications for Statistical Applications and Methods
• /
• 제5권2호
• /
• pp.503-515
• /
• 1998

When a neutral particle beam(NPB) aimed at the object and receive a small number of neutron signals at the detector, it follows approximately Poisson distribution. Under the four assumptions in the presence of errors and uncertainties for the Poisson parameters, an exact probability distribution of neutral particles have been derived. The probability distribution for the neutron signals received by a detector averaged over the three sources of errors is expressed as a four-dimensional integral of certain data. Two of the four integrals can be evaluated analytically and thereby the integral is reduced to a two-dimensional integral. The monotone likelihood ratio(MLR) property of the distribution is proved by using the Cauchy mean value theorem for the univariate distribution and multivariate distribution. Its MLR property can be used to find a criteria for the hypothesis testing problem related to the distribution.

• 대한수학회지
• /
• 제47권2호
• /
• pp.385-407
• /
• 2010

To study the relationship between the levels of chemical pollutants and the number of daily total hospital admissions for respiratory diseases and to find the effect of temperature/relative humidity on the admission number, Wong et al. [17] introduced the varying-coefficient single-index model (VCSIM). As pointed out, it is a popular multivariate nonparametric fitting technique. However, the tests of the model have not been very well developed. In this paper, based on the estimators obtained by the local linear technique, the average method and the one-step back-fitting technique in the VCSIM, the generalized likelihood ratio (GLR) tests for varying-coefficient parts on the VCSIM are established. Under the null hypotheses the new proposed GLR tests follow the $\chi^2$-distribution asymptotically with scale constant and degree of freedom independent of the nuisance parameters, known as Wilks phenomenon. Simulations are conducted to evaluate the test procedure empirically. A real example is used to illustrate the performance of the testing approach.

• 대한수학회논문집
• /
• 제20권1호
• /
• 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 applied mathematics & informatics
• /
• 제6권1호
• /
• pp.31-50
• /
• 1999

We revisit the problems of testing three-factor classifica-tion models with a single observation per cell. A common approach in analyzing such nonreplicated data is to omit the highest order in-teraction and regard it as error. This paper discusses the use of a multiplicative model(See and Smith 1996 and 1998) which is applied on residuals in order to separate the variablility due to three-factor interaction from what is counted as random error. in particualr to test the significance of the interaction term we derived an approxi-mated distribution of the likelihood ratio test statistic based on the quadrilinear model known as Tucher's three-mode principal compo-nent model. The derivation utilizes the distribution of the eignevalues of the Wishart matrix.

• International Journal of Reliability and Applications
• /
• 제15권1호
• /
• pp.65-76
• /
• 2014

In this article, a new model based on Lomax distribution is introduced. This new model is both useful and practical in areas such as economic, reliability and life testing. Some statistical properties of this model are presented including moments, hazard rate, reversed hazard rate, mean residual life and mean inactivity time functions, among others. It is also shown that the distributions of the new model are ordered with respect to the strongest likelihood ratio ordering. The method of moment and maximum likelihood estimation are used to estimates the unknown parameters. Simulation is utilized to calculate the unknown shape parameter and to study its properties. Finally, to illustrate the concepts, the appropriateness of the new model for real data sets are included.

• International Journal of Reliability and Applications
• /
• 제14권1호
• /
• pp.27-39
• /
• 2013

In this article, a new model based on the Rayleigh distribution is introduced. This model is useful and practical in physics, reliability, and life testing. The statistical and reliability properties of this model are presented, including moments, the hazard rate, the reversed hazard rate, and mean residual life functions, among others. In addition, it is shown that the distributions of the new model are ordered regarding the strongest likelihood ratio ordering. Four estimating methods, namely, method of moment, maximum likelihood method, Bayes estimation, and uniformly minimum variance unbiased, are used to estimate the parameters of this model. Simulation is used to calculate the estimates and to study their properties. Finally, the appropriateness of this model for real data sets is shown by using the chi-square goodness of fit test and the Kolmogorov-Smirnov statistic.

• Communications for Statistical Applications and Methods
• /
• 제2권1호
• /
• pp.112-121
• /
• 1995

Let $p_1$ and $p_2$ be the proportions of two populations. To test the hypothesis $H_0 : p_1 = p_2$, we usually use the $x^2$ statistic, the large sample binomial statistic Z, and the Generalized Likelihood Ratio statistic-2log $\lambda$developed based on different mathematical rationale, respectively. Since testing the above hypothesis is equivalent to testing whether two populations follow the common Bernoulli distribution, one may also test the hypothesis by comparing 1 with the ratio of each density estimate and the hypothesized common density estimate, called comparison density, which was devised by Parzen(1988). We show that the above traditional test statistics ate actually estimating the measure of distance between the true densities and the common density under $H_0$ by representing them with the comparison density.