• Journal of the Korean Data and Information Science Society
• /
• v.24 no.3
• /
• pp.477-485
• /
• 2013

In this paper, we consider the sample sizes required for each group in independent two sample test of normal populations when both the type I error and type II error probabilities are specified with sample sizes and variances being possibly different. We derived the sample sizes and the power of the tests, and implement them by web programing. The result is available over the world wide web. Further, we also provide the power calculations and have them available on the web.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.1
• /
• pp.53-61
• /
• 2012

There are cases when the sample size is determined based not only on the significance level but also on on the power or type II error. In this paper, we implemented the sample size and the power calculation when both the significance level and power for testing means in normal distributions and proportions in binomial distributions. The implementation is available on a web site. Alternately, we also calculate the power for a given effect size, type I error probability and sample size.

• The Korean Journal of Applied Statistics
• /
• v.12 no.2
• /
• pp.593-603
• /
• 1999

표본의 크기의 제1종오류의 확률 $\alpha$, 실용적으로 차이가 있다고 판독되어서 검출하고자하는 요인효과의 오차에 대한 상대적인 크기, 그 값에서의 제2종오류의 확률 $\beta$에 따라서 결정된다. 이 논문에서, 우리는 고정요인과 랜덤요인이 포함된 실험계획에서 표본의 크기를 결정하는 방법을 간단한 MATLAB 프로그램을 사용하여 고려한다. 분할법과 지분요인배치법의 예제를 들어 유의수준 $\alpha$와 최소 표준과 검출효과 $\Delta^*$에서 검정력이 적어도 $1-\beta$를 갖도록 표본의 크기를 결정한다

• Communications for Statistical Applications and Methods
• /
• v.17 no.1
• /
• pp.127-140
• /
• 2010

In interim analysis, group sequential tests are widely used for the ethical, scientific, and economic reasons. In this paper, we propose the group sequential tests using both type I and type II error spending rate functions when the response variable is discrete, especially binomial distribution, in the interim analysis. In addition, we propose new error spending rate function which covers the formerly proposed. Our method has good property that is flexible, fast and easily applicable. A numerical simulations are carried out to evaluate our method and it shows good performance.

• The Korean Journal of Applied Statistics
• /
• v.33 no.4
• /
• pp.395-407
• /
• 2020

Most classification accuracy measures for optimal threshold are divided into two types: one is expressed with cumulative distribution functions and probability density functions, the other is based on ROC curve and AUC. Unal (2017) proposed the index of union (IU) as an accuracy measure that considers two types to get them. In this study, ten kinds of accuracy measures (including IU) are divided into six categories, and the advantages of the IU are studied by comparing the measures belonging to each category. The optimal thresholds of these measures are obtained by setting various normal mixture distributions; subsequently, the first and second type of errors as well as the error sums corresponding to each threshold are calculated. The properties and characteristics of the IU statistic are explored by comparing the discriminative power of other accuracy measures based on error values.The values of the first type error and error sum of IU statistic converge to those of the best accuracy measures of the second category as the mean difference between the two distributions increases. Therefore, IU could be an accuracy measure to evaluate the discriminant power of a model.

• 한국데이터정보과학회:학술대회논문집
• /
• 2004.10a
• /
• pp.17-23
• /
• 2004

At the conclusion from the hypothesis testing, there is a possibility of making Type I error and Type II error. The purpose of this article is to use this program in statistics teaching through developing the program for studying on the concept about these two errors, two kinds of the probability of errors by the variation of rejection region, two kinds of the probability of errors by the variation of sample size, the relations of the probability $\alpha$ and $\beta$ by these two errors, and power function, power curve.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.1
• /
• pp.19-30
• /
• 2006

The purpose of this article is to develop a program for teaching Type I and Type II errors in one sample. In this program, the concepts of two errors are visually explained, and the probabilities of two errors are also visually displayed as the rejection region changes or the sample size changes. Also, in this program, the power curve and the operating characteristic curve are visually displayed in terms of the parameter value of interest.

• Journal of The Korean Society For Urban Railway
• /
• v.6 no.4
• /
• pp.401-410
• /
• 2018

Human error of the engineer are constantly occurring by insufficiency of scientific analysis considering the behavior of railway engagements and actual working conditions and lack of research for accidents. This study investigated how to improve the system through AHP and FGI analysis of interviews and questionnaires for electric vehicles driving licenses, which cultivates the most train drivers.

• The Korean Journal of Applied Statistics
• /
• v.8 no.2
• /
• pp.87-97
• /
• 1995

When there is no main effcts or only one main effect in a $2 \times 2$ factorial design, Type I error rates and power for the rank transformed F test (FR test) for interaction are nearly equal to those of the classical F test. However the power of FR test is quite superior under the exponential distribution rather than the of FR test is quite superior under the exponential distribution rather than the normal distribution. Meanwhile when both main effects are in the model, Type I error rates of FR test, compared with those of F test, decrease as the effect size increases and are dependent on the fashion in which main effects are constructed. In addition, the power of FR test increases as the effect size and the sample size increase and is highly dependent on the manner in which main effects are constructed and the type of population distribution.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.5
• /
• pp.1153-1165
• /
• 2017

Structural equation modeling (SEM) is a basic tool for testing theories in a variety of disciplines. A maximum likelihood (ML) method for parameter estimation is by far the most widely used in SEM. Alternatively, two-stage least squares (2SLS) estimator has been proposed as a more robust procedure to address model misspecification. A regularized extension of 2SLS, two-stage ridge least squares (2SRLS) has recently been introduced as an alternative to ML to effectively handle the small-sample-size issue. However, it is unclear whether and when misspecification and small sample sizes may pose problems in theory testing with 2SLS, 2SRLS, and ML. The purpose of this article is to evaluate the three estimation methods in terms of inferences errors as well as parameter recovery under two experimental conditions. We find that: 1) when the model is misspecified, 2SRLS tends to recover parameters better than the other two estimation methods; 2) Regardless of specification errors, 2SRLS produces small or relatively acceptable Type II error rates for the small sample sizes.