• The Korean Journal of Applied Statistics
• /
• v.26 no.4
• /
• pp.595-610
• /
• 2013

Testing of order-restricted alternative hypothesis in $2{\times}k$ contingency tables can be applied to various fields of medicine, sociology, and business administration. Most testing methods have been developed based on a large sample theory. In the case of a small sample size or unbalanced sample size, the Type I error rate of the testing method (based on a large sample theory) is very different from the target point of 5%. In this paper, the exact testing method is introduced in regards to the testing of an order-restricted alternative hypothesis in categorical data (particularly if a small sample size or extreme unbalanced data). Power and exact p-value are calculated, respectively.

• Journal of the Korea Safety Management & Science
• /
• v.15 no.1
• /
• pp.283-292
• /
• 2013

This paper aims to propose a new steps of hypothesis testing using analysis process and improvement process in the six sigma DMAIC. The six sigma implementation models proposed in this paper consist of six steps. The first step is to establish a research hypothesis by specification directionality and FBP(Falsibility By Popper). The second step is to translate the research hypothesis such as RHAT(Research Hypothesis Absent Type) and RHPT(Research Hypothesis Present Type) into statistical hypothesis such as $H_0$(Null Hypothesis) and $H_1$(Alternative Hypothesis). The third step is to implement statistical hypothesis testing by PBC(Proof By Contradiction) and proper sample size. The fourth step is to interpret the result of statistical hypothesis test. The fifth step is to establish the best conditions of product and process conditions by experimental optimization and interval estimation. The sixth step is to draw a conclusion by considering practical significance and statistical significance. Important for both quality practitioners and academicians, case analysis on six sigma projects with implementation guidelines are provided.

• Journal of the Korean Statistical Society
• /
• v.23 no.1
• /
• pp.199-206
• /
• 1994

Sometimes a researcher with a view of conventional significance testing rejects his/her hypothesis, even through it could have not been rejected with a smaller sample. This can be a logical dilemma for a researcher who wants to "prove" a hypothesis rather than to show discrepancy from a null hypothesis. In this study, a new testing paradigm called equivalence testing via confidence interval will be developed so that it is suitable for the purpose of statistical proof.cal proof.

• Journal of the Korean Statistical Society
• /
• v.35 no.2
• /
• pp.167-177
• /
• 2006

An important role of statistical analysis in science is interpreting observed data as evidence, that is 'what do the data say?'. Although standard statistical methods (hypothesis testing, estimation, confidence intervals) are routinely used for this purpose, the theory behind those methods contains no defined concept of evidence and no answer to the basic question 'when is it correct to say that a given body of data represent evidence supporting one statistical hypothesis against another?' (Royall, 1997). In this article, we use likelihood ratios to measure evidence provided by record values in favor of a hypothesis and against an alternative. This hypothesis is concerned on mean of an exponential model and prediction of future record values.

• Korean Business Review
• /
• v.18 no.1
• /
• pp.19-41
• /
• 2005

The informativeness of accounting earnings to investors may be used as a measure of the quality of accounting information. The ownership structure is considered to be related to both of two variables, the reliability of earnings reflecting the economic performance of the enterprise and the existence of alternative information sources. Earnings manipulation may mitigate the correlation between accounting earnings and the economic performance and alternative information sources decrease the value of accounting earnings as an information. Thus ownership structure could influence the informativeness of accounting earnings. This paper classifies ownership into three categories, management or inside ownership, institutional investors and large outside blockholders, and diffuse outside ownership and examines theoretically the difference of information effectiveness under each ownership structure. The earnings manipulation hypothesis supports the assertion that the separation of ownership from control motivates earnings manipulation. And differential information hypothesis suggests that more non-accounting information of firms with institutional or concentrated outside ownership is provided. Outside blockholders have alternative information sources that make accounting manipulation ineffective. While most previous studies have examined the effect of ownership on the informativeness of earnings from earnings manipulation hypothesis, this study is motivated by both earnings manipulation hypothesis and differential information hypothesis.

• Journal of the Korean Statistical Society
• /
• v.36 no.2
• /
• pp.201-208
• /
• 2007

Evolutionary operation (EVOP) proposed by Box (1957) is a method for continuous monitoring and improvement of a full-scale manufacturing process with the objective of moving the operating conditions toward the better ones. EVOP consists of systematically making small changes in the levels of the two or three process variables under consideration. Data are collected on the response variable at each point of two level factorial design with the center point and a cycle is said to have been completed. The cycles are replicated sequentially until the decision is made on whether further cycle of experiments is needed to conclude the significance of any of main effects or interaction effects or the curvature. In this paper, an improved flow chart of EVOP is proposed and how to determine the number of cycles is studied based on the size of type II error. In order to reject the alternative hypothesis of interests with more confidence and conclude that we believe in the null hypothesis of no effects, we propose a counter measure $p^*-value$ corresponding to the p-value. The relationship of $p^*-value$ to the probability of type II error ${\beta}$ under the alternative hypothesis of interests is analogous to that of p-value to the probability of type I error ${\alpha}$. Also the implementation of EVOP with a mixture experiment is discussed.

• Interdisciplinary Bio Central
• /
• v.1 no.3
• /
• pp.11.1-11.5
• /
• 2009

One of the main issues of molecular evolution is to reveal the principles dictating protein evolutionary rates. A traditional hypothesis posits that protein evolutionary rates are mostly determined by the average functional importance of amino acids in a given protein. Thus the correlations of evolutionary rates with different variables such as PPI, gene essentiality and expression abundance have been studied to test the traditional hypothesis. Recently, mRNA expression abundance among the variables has drawn much attention, not only because it shows relatively strong correlation with protein evolutionary rates, but also because of the controversies surrounding an alternative hypothesis against the traditional one. Here, I will give an overview over the traditional hypothesis, and summarize the different variables that have been found to correlate with protein evolutionary rates. Then I will introduce pros and cons on the two different hypotheses.

• Proceedings of the KSPS conference
• /
• 2005.04a
• /
• pp.95-98
• /
• 2005

Many utterance verification (UV) algorithms have been studied to reject out-of-vocabulary (OOV) in speech recognition systems. Most of conventional confidence measures for UV algorithms are mainly based on log likelihood ratio test, but these measures take much time to evaluate the alternative hypothesis or anti-model likelihood. We propose a novel confidence measure which makes use of a momentary best scored state sequence during Viterbi search. Our approach is more efficient than conventional LRT-based algorithms because it does not need to build anti-model or to calculate the alternative hypothesis. The proposed confidence measure shows better performance in additive noise-corrupted speech as well as clean speech.

• ETRI Journal
• /
• v.18 no.3
• /
• pp.207-213
• /
• 1996

This paper proposes a new nonparametric test for testing the null hypothesis that the MRL is constant against the alternative hypothesis that the MRL is decreasing (increasing) for ramdomly censored data. The proposed test statistic is a L-statistic, and we use L-statistic theory to establish its asymptotic normality of the test statistic. We discuss the efficiency loss due to censoring and also calculate the asymptotic relative efficiencies of our test statistic with respect to the Chen, Hollander and Langberg's test for several alternatives.

• Journal of the Korean Chemical Society
• /
• v.51 no.3
• /
• pp.279-286
• /
• 2007

In this study, we investigated the influences of cognitive conflict and situational interest induced by a discrepant event and an alternative hypothesis, attention and state learning strategies on conceptual change. A preconception test was administered to 486 seventh graders. They also completed the questionnaires of cognitive response and situational interest to a discrepant event before/after presenting an alternative hypothesis. After learning the concept of density with a CAI program as conceptual change intervention, the tests of attention, state learning strategies, and conceptual understanding were administered as posttests. Analyses of the results for 197 students having misconceptions about density revealed that post-cognitive conflict was significantly higher than pre-cognitive conflict. However, there was no statistically significant difference between the test scores of pre-situational interest and post-situational interest. Pre-cognitive conflict only exerted a direct effect on post-cognitive conflict, while post-cognitive conflict exerted a direct effect and Journal of the Korean Chemical Society an indirect effect via attention on conceptual understanding. Both pre- and post-situational interests were found to influence on conceptual understanding via attention. Attention had influences positively on deep learning strategy and negatively on surface learning strategy. There was a relatively small effect of state learning strategies on conceptual understanding.