• Journal of Integrative Natural Science
• /
• v.12 no.3
• /
• pp.91-99
• /
• 2019

Many researchers in various study fields use the two sample t-test to confirm their treatment effects. The two sample t-test is generally used for small samples, and assumes that two independent random samples are selected from normal populations, and the population variances are unknown. Researchers often conduct F-test, the test of equality of variances, before testing the treatment effects, and the test statistic or confidence interval for the two sample t-test has two formats according to whether the variances are equal or not. Researchers using the two sample t-test often want to know how large sample sizes they need to get reliable test results. This research gives some guidelines for sample sizes to them through simulation works. The simulation had run for normal populations with the different ratios of two variances for different sample sizes (${\leq}30$). The simulation results are as follows. First, if one has no idea equality of variances but he/she can assume the difference is moderate, it is safe to use sample size at least 20 in terms of the nominal level of significance. Second, the power of F-test for the equality of variances is very low when the sample sizes are small (<30) even though the ratio of two variances is equal to 2. Third, the sample sizes at least 10 for the two sample t-test are recommendable in terms of the nominal level of significance and the error limit.

• Communications for Statistical Applications and Methods
• /
• v.17 no.1
• /
• pp.79-88
• /
• 2010

Introductory statistic class proposes two tests for the equality of two population means according to the homogeneity of their variances. However, in practice, the variances are also unknown and practitioners often test their homogeneity before they do two sample t-test. This is also true in many popular statistical packages such as SAS and SPSS. In this paper, we study the type I error of this two stage procedure and propose a procedure to control it at a given significance level.

• Journal of the Korean Data and Information Science Society
• /
• v.16 no.3
• /
• pp.609-620
• /
• 2005

In the secondary data analysis for categorical data, situations often arise in which the estimated cell variances are available, but not the full matrix of variances. In this case researchers are often inclined to use Pearson-type test statistics for homogeneity. However, for a complex sample observed cell proportions are not distributed as multinomial and Pearson-type test statistic generally is not distributed asymptotically as chi-square distribution. This paper evaluates powers for Wald test and Pearson-type test and the first order corrected test of Pearson-type test for homogeneity. The resulting power curves indicate that as the misspecification effect increases, the amount of inflation of significance level and the loss of power Pearson-type test are getting more severe.

• Communications for Statistical Applications and Methods
• /
• v.9 no.3
• /
• pp.853-864
• /
• 2002

In the analysis of categorical data subject to misclassification errors, the observed cell proportions are adjusted by a misclassification probabilities and estimates of variances are adjusted accordingly. In this case, it is important to determine the extent to which misclassification probabilities are homogeneous within a population. This paper considers methods to evaluate the power of chi-squared tests for homogeneity with complex survey data subject to misclassification errors. Two cases are considered: adjustment with homogeneous misclassification probabilities; adjustment with heterogeneous misclassification probabilities. To estimate misclassification probabilities, logistic regression method is considered.

• Communications for Statistical Applications and Methods
• /
• v.16 no.2
• /
• pp.389-396
• /
• 2009

Approximations for the null distribution of a test statistic arising in multivariate analysis to test homogeneity of variances and a mixture of two beta distributions by making use of a product of beta baseline density function and a polynomial adjustment, so called beta-polynomial density approximant, are discussed. Explicit representations of density and distribution approximants of interest in each case can easily be obtained. Beta-polynomial density approximants produce good approximation over the entire range of the test statistic and also accommodate even the bimodal distribution using an artificial example of a mixture of two beta distributions.

• Journal of Integrative Natural Science
• /
• v.7 no.3
• /
• 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.

• Journal of the Korean Data and Information Science Society
• /
• v.16 no.2
• /
• pp.195-205
• /
• 2005

In decision tree analysis, C4.5 and CART algorithm have some problems of computational complexity and bias on variable selection. But QUEST algorithm solves these problems by dividing the step of variable selection and split point selection. When input variables are continuous, QUEST algorithm uses ANOVA F-test under the assumption of normality and homogeneity of variances. In this paper, we investigate the influence of violation of normality assumption and effect of the transformation of variables in the QUEST algorithm. In the simulation study, we obtained the empirical powers of variable selection and the empirical bias of variable selection after transformation of variables having various type of underlying distributions.

• Communications for Statistical Applications and Methods
• /
• v.31 no.1
• /
• pp.1-35
• /
• 2024

The assumption of homoscedasticity is one of the most crucial assumptions for many parametric tests used in the biological sciences. The aim of this paper is to compare the empirical probability of type I error and the power of ten parametric and two non-parametric tests for homoscedasticity with simulations under different types of distributions, number of groups, number of samples per group, variance ratio and significance levels, as well as through empirical data from an agricultural experiment. According to the findings of the simulation study, when there is no violation of the assumption of normality and the groups have equal variances and equal number of samples, the Bhandary-Dai, Cochran's C, Hartley's Fmax, Levene (trimmed mean) and Bartlett tests are considered robust. The Levene (absolute and square deviations) tests show a high probability of type I error in a small number of samples, which increases as the number of groups rises. When data groups display a nonnormal distribution, researchers should utilize the Levene (trimmed mean), O'Brien and Brown-Forsythe tests. On the other hand, if the assumption of normality is not violated but diagnostic plots indicate unequal variances between groups, researchers are advised to use the Bartlett, Z-variance, Bhandary-Dai and Levene (trimmed mean) tests. Assessing the tests being considered, the test that stands out as the most well-rounded choice is the Levene's test (trimmed mean), which provides satisfactory type I error control and relatively high power. According to the findings of the study and for the scenarios considered, the two non-parametric tests are not recommended. In conclusion, it is suggested to initially check for normality and consider the number of samples per group before choosing the most appropriate test for homoscedasticity.

• Journal of Biomedical Engineering Research
• /
• v.19 no.6
• /
• pp.551-562
• /
• 1998

Amultiresolution filtering algorithm based on the physical SPECT lesion detachability provides and optimal solution for SPECT reconstruction problem. Related to the previous study, we estimated the SPECT lesion detection capability by m minimum detectable lesion sizes (MDLSs), and generated m reconstruction filters which are designed to maximize the smoothing effect at a fixed MDLS-dependent resolution level $\frac{MDLS}{4\sqrt{2In2}}$. The proposed multiresolution filtering algorithm used a coarse-to-fine approach for the m-level resolution filter images obtained from these m filters for a given projection image. First, the local homogeneity is determined for every pixel of the filter images, by comparing the local variance value computed in a window centered at the pixel and the mode determined from the distribution of the local variances. Based on the local homogeneity, the pixels declared as homogeneous are chosen from the filter image of the lowest resolution, and for the other pixels the same process is repeated for the higher resolution filter images. For the non-homogeneous pixels after this pixels after this repetition process ends, the pixel values of the highest resolution filter image are substituted. From the results of the simulated experiments, the proposed multiresolution filtering algorithm showed a strong smoothing effect in the homogeneous regions and a significant resolution improvement near the edge regions of the projection images, and so produced good adaptability effects in the reconstructed images.

• Journal of Acupuncture Research
• /
• v.37 no.2
• /
• pp.110-117
• /
• 2020

Background: The aim of this pilot study was to assess the safety and dosing of scolopendrid pharmacopuncture (SPP). Methods: A total of 40 healthy Sprague-Dawley rats (males and 20 females 20) were selected following a 7-day inspection and acclimation period. SPP was administered via intramuscular injection, over a 2-week period using 3 doses including a high-dose [0.84 mg of scolopendrid per kg of body weight (BW)], a med-dose (0.42 mg/kg BW), and a low-dose (0.21 mg/kg BW). The control group was injected with sterile water into the muscles. Unusual changes caused by administration of the test substance were observed. Weight, feed intake, organ weight, and hematological examinations were compared among the groups. Using the SPSS statistical program, Levene's test was performed to evaluate the homogeneity of variances, and a one-way ANOVA test was subsequently performed to assess the significance between each test group. Results: During the experiment no animals died. Weight change, food consumption, organ weight, hematological test, and blood biochemical tests showed no significant differences in the treatment groups compared to controls. Conclusion: No toxicological changes related to the administration of test substances were observed. Therefore, the LD₅₀ (lethal-dose that kills 50%) of scolopendrid pharmacoupuncture in rats was greater than 0.84 mg/kg.