• Communications for Statistical Applications and Methods
• /
• 제6권1호
• /
• pp.11-24
• /
• 1999

In this paper we propose a method to test $\textit{H}$:$\rho_i$=$\gamma_i$ for 1$\leq$$\textit{i}$$\leq$$\ell$ against $\textit{K}$:$\rho_i$$\neq$$\gamma_i$ for some ｉin k-variance component random or mixed linear model where $\rho$i denotes the ratio of the i-th variance component to the error variance and $\ell$$\leq$K. The test which we call Rao-Wald test is exact and does not depend upon nuisance parameters. From a numerical study of the power performance of the test of the interaction effect for the case of a two-way random model Rao-Wald test was seen to be quite comparable to the locally best invariant (LBI) test when the nuisance parameters of the LBI test are assumed known. When the nuisance parameters of the LBI test are replaced by maximum likelihood estimators Rao-Wald test outperformed the LBI test.

• Journal of the Korean Data and Information Science Society
• /
• 제16권1호
• /
• pp.137-144
• /
• 2005

In $I{\times}J$ incomplete contingency tables, the test of independence proposed by Chen and Fienberg(1974) uses $I{\times}J-1$ instead of (I-1)(J-1) degrees of freedom without providing much of an increase in the value of the test statistic. For these reasons, Chen and Fienberg tests are expected to have less power. New Wald test statistic related to the part of Chen and Fienberg test statistic is proposed using delta method. These two tests are compared through Monte Carlo studies.

• Journal of the Korean Statistical Society
• /
• 제32권2호
• /
• pp.93-101
• /
• 2003

This paper deals with testing the equality of the coefficients of variation in two inverse Gaussian populations. The likelihood ratio, Lagrange-multiplier and Wald tests are presented. Monte-Carlo simulations are performed to compare the powers of these tests. In a simulation study, the likelihood ratio test appears to be consistently more powerful than the Lagrange-multiplier and Wald tests when sample size is small. The powers of all the tests tend to be similar when sample size increases.

• International Journal of Control, Automation, and Systems
• /
• 제4권3호
• /
• pp.333-343
• /
• 2006

In this paper, the baseline-length information is directly modeled as a measurement for the Wald test, which speeds up the resolution convergence of the integer ambiguity of GPS carrier phase measurements. The convergent speed improvement is demonstrated using numerical simulation and real experiments. It is also shown that the integer ambiguities can be resolved using only four actual satellite measurements with very reasonable convergence speed, if the baseline-length information is used just like one additional observable satellite measurement. Finally, it is shown that the improvement of convergence speed of the Wald test is due to the increase of the probability ratio with the use of the baseline-length constraint.

• International Journal of Control, Automation, and Systems
• /
• 제6권4호
• /
• pp.534-543
• /
• 2008

To use the Global Positioning System (GPS) carrier phase measurement for precise positioning, the integer ambiguities at the early stage of most algorithms must be determined. Furthermore, if a precise positioning is to be applied to real time navigation, fast determination and validation methods for integer ambiguity are essential. In this paper, the Wald test that simultaneously determines and validates integer ambiguities is used with assistance of the Inertial Navigation System (INS) to improve its performance. As the Wald test proceeds, it assigns a higher probability to the candidate that is considered to be true at each time step. The INS information is added during the Wald test process. Large performance improvements were achieved in convergence time as well as in requiring fewer observable GPS satellites. To test the performance improvement of the Wald test with the INS information, experimental tests were conducted using a ground vehicle. The vehicle moved in a prescribed trajectory and observed seven GPS satellites. To verify the effect of the INS information on the Wald test, the convergence times were compared with cases that considered the INS information and cases that did not consider the INS information. The results show that the benefits of using the INS were emphasized as fewer GPS satellites were observable. The performance improvement obtained by the proposed algorithm was shown through the fast convergence to the true hypothesis when using the INS measurements.

• 통합자연과학논문집
• /
• 제7권3호
• /
• pp.208-213
• /
• 2014

National-wide and/or large scale sample surveys generally use complex sample design. Traditional Pearson chi-square test is not appropriate for the categorical complex sample data. Rao-Scott suggested an adjustment method for Pearson chi-square test, which uses the average of eigenvalues of design matrix of cell probabilities. This study is to compare the efficiency of Rao-Scott first order adjusted test to Wald test for homogeneity between two populations using 2009 Gyeongnam regional education offices's customer satisfaction survey (2009 GREOCSS) data. The 2009 GREOCSS data were collected based on stratified three-stage cluster sampling with probability proportional to size. The empirical results show that the Rao-Scott adjusted test statistic using only the variances of cell probabilities is very close to the Wald test statistic, which uses the covariance matrix of cell probabilities, under the 2009 GREOCSS data based. However it is necessary to be cautious to use the Rao-Scott first order adjusted test statistic in the place of Wald test because its efficiency is decreasing as the relative variance of eigenvalues of the design matrix of cell probabilities is increasing, specially more when the number of degrees of freedom is small.

• 응용통계연구
• /
• 제29권1호
• /
• pp.123-132
• /
• 2016

패널 시계열 자료를 소개하고 패널 1차 자기회귀 모형을 고려하였다. 패널 1차 자기회귀 모형의 동질성 검정을 위한 검정 통계량으로 Rao 통계량과 Wald 통계량을 제안하고, 그 극한분포를 제시하였다. 모의실험을 통해 패널의 수가 작을 때에도 패널의 수가 많을 때와 마찬가지로 두 검정 통계량의 분포가 카이제곱분포를 따르는 것을 확인하였으며, 패널의 수가 작을 때 Rao 통계량이 Wald 통계량 보다 더 우수한 검정력을 가짐을 모의실험을 통해 확인하였다. 시도별 월별 경제활동인구수 자료를 패널 1차 자기회귀 모형으로 적합하여 동질성 검정을 수행한 결과 동질성을 만족하였다. 동질성 검정을 만족한 자료를 시점별 평균을 이용하여 종합하고 이를 1차 자기회귀모형으로 적합하였다. 각각의 시도별로 적합한 모형과 시점별 평균을 이용하여 적합한 모형의 예측력을 비교한 결과 동질성 검정을 통과한 패널 1차 자기회귀모형의 경우 자료를 종합하여 적합한 모형의 예측력이 더 우수함을 확인하였다.

• 한국전자통신학회논문지
• /
• 제5권5호
• /
• pp.435-443
• /
• 2010

확률적 특성을 가지는 시스템의 시험을 위해서는 시험 입력을 일정 횟수만큼 반복하여 제공하고 관찰된 데이터를 기반으로 판정이 내려져야 한다. 구간 추정 기법을 이용하여 관찰된 데이터로부터 확률 값이 올바른지 여부를 판단할 수 있으며, 이 때 적절한 신뢰구간의 선택은 시험의 품질을 결정하는 중요한 요인이 된다. 본 논문에서는 다양한 크기의 표본에 대해 대표적인 구간 추정 기법인 Wald 신뢰구간과 Agresti-Coull 신뢰구간을 비교 분석한다. 각 신뢰구간이 확률 값 시험에 사용되었을 경우 올바른 구현 제품이 시험을 통과할 확률과 잘못된 구현제품이 시험을 통과하지 못할 확률을 기반으로 비교 분석을 수행하며, 확률 값이 올바른지를 판단하기 위한 양측검정뿐만 아니라 확률 값이 기준 확률 이상인지 여부를 판단하기 위한 단측검정을 사용하는 경우에 대해서도 비교 분석을 수행한다. 비교 분석 결과 양측검정의 경우 Agresti-Coull 신뢰구간을 사용할 것을 추천하며, 단측검정의 경우 큰 크기의 표본에 대해서는 Agresti-Coull 신뢰구간을, 적은 크기의 표본에 대해서는 Wald 신뢰구간 또는 Agresti-Coull 신뢰구간을 선택적으로 사용할 것을 추천한다.

• Communications for Statistical Applications and Methods
• /
• 제11권3호
• /
• 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.

• 응용통계연구
• /
• 제14권1호
• /
• pp.211-221
• /
• 2001

본 논문에서는 2$\times$2 분할표의 주변동질성 검정에서 사용될 수 있는 통계량들을 소개하고, 이 통계량들을 비교하였다. 먼저 주변동질성 검정에 민감하게 영향을 주는 모수를 정의한 후에 이 모수들의 효과를 예시하였다. 또한 이 모수들을 이용하여 모의실험을 통해여러 검정법들을 비교해본 결과 McNemar 검정이 다른 검정력보다 더 좋은 성질을 가지고 있음을 보였다.