• 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.

• 대한교통학회지
• /
• 제13권4호
• /
• pp.61-77
• /
• 1995

The main objective of this research is the development of a logit model of urban workers' auto ownership choice. For the utility specification. a variety of behavioral hypotheses about the factors which affect the urban workers' auto ownership choice are considered. Based on the behavioral hypotheses, a binary logit model of auto ownership is estimated. Empirical estimation is based on a sample of workers taken in Daegu City(1994). The binary logit model of auto ownership development in this paper provides reasonable results in terms of behavioral and statistical considerations. Furthermore, this paper develops several submarket models of auto ownership choice. Market segmentation was made using age, sex, income, home-to-work time distance. It is found that the estimated results with market segmentation are also reasonable. Finally future directions of model development are suggested.

• Journal of the Korean Data and Information Science Society
• /
• 제16권4호
• /
• pp.1129-1140
• /
• 2005

A nonparametric Bayesian method for calculating posterior probabilities of the multiple comparison problem on the parameters of several Geometric populations is presented. Bayesian multiple comparisons under two different prior/ likelihood combinations was studied by Gopalan and Berry(1998) using Dirichlet process priors. In this paper, we followed the same approach to calculate posterior probabilities for various hypotheses in a statistical experiment with a partition on the parameter space induced by equality and inequality relationships on the parameters of several geometric populations. This also leads to a simple method for obtaining pairwise comparisons of probability of successes. Gibbs sampling technique was used to evaluate the posterior probabilities of all possible hypotheses that are analytically intractable. A numerical example is given to illustrate the procedure.

• 한국산업정보학회:학술대회논문집
• /
• 한국산업정보학회 2008년도 추계 공동 국제학술대회
• /
• pp.194-198
• /
• 2008

Under the rapid change of business circumstance to E-business, CRM is one of the major strategic choices to reinforce the relationship with customers. In this point, this study attempts to understand the relationships among the quality of CRM and the variables of post adoption on the innovation diffusion theory. With integration of information system success model and innovation diffusion theory, it developed a research model and several hypotheses focusing on CRM performance of post-adoption. To testify the hypotheses, it conducted a survey on insurance industry and performed a statistical analysis for the research model with structure equation model. The results show that the factors of CRM quality (system, service, and information) have significant relationships with level of CRM usage and satisfaction. And the relationships between satisfaction and CRM performance and between CRM performance and intension of CRM extension are found as statistically significant.

• Journal of the Korean Data and Information Science Society
• /
• 제10권1호
• /
• pp.135-146
• /
• 1999

In this paper, we consider the Bayesian hypotheses testing for independence and symmetry in Freund's bivariate exponential model. In Bayesian testing problem, we use the noninformative priors for parameters which are improper and are defined only up to arbitrary constants. And we use the recently proposed hypotheses testing criterion called the intrinsic Bayes factor. Also we derive the arithmetic and median intrinsic Bayes factors and use these results to analyze some data sets.

• 한국정보시스템학회지:정보시스템연구
• /
• 제15권4호
• /
• pp.55-72
• /
• 2006

The total market capacities of our domestic on-line shopping malls had reached over one billion wons in 2005. They are also estimated to reach 1.8 billion wons in 2008. In order to reveal some relationships between six antecedent variables and intention-to-use of online shooing malls, this study has extended the original TAM2 model using these variables obtained from (1) perceived usefulness and perceived ease of use from TAM2 model, (2) compatibility from IDT theory, (3) perceived enjoyment from Flow theory, and (4) others (perceived risk and self efficacy). Two statistical packages such as SPSS 12.0 and Lirel 8.70 were used for data analyses. Among these seven proposed hypotheses, six hypotheses were accepted while one hypothesis regarding perceived risk was rejected. As perceived risk hypothesis concerning the intention-to-use was rejected, the perceived risk did not show the support of the intention-to-use.

• 음성과학
• /
• 제11권4호
• /
• pp.43-52
• /
• 2004

Model parameters in HMM based speech recognition systems are normally estimated using Maximum Likelihood Estimation(MLE). The MLE method is based mainly on the principle of statistical data fitting in terms of increasing the HMM likelihood. The optimality of this training criterion is conditioned on the availability of infinite amount of training data and the correct choice of model. However, in practice, neither of these conditions is satisfied. In this paper, we propose a training algorithm, MCE(Minimum Classification Error), to improve the performance of a speech recognizer detecting mispronunciation of a foreign language. During the conventional MLE(Maximum Likelihood Estimation) training, the model parameters are adjusted to increase the likelihood of the word strings corresponding to the training utterances without taking account of the probability of other possible word strings. In contrast to MLE, the MCE training scheme takes account of possible competing word hypotheses and tries to reduce the probability of incorrect hypotheses. The discriminant training method using MCE shows better recognition results than the MLE method does.

• 복식문화연구
• /
• 제20권3호
• /
• pp.443-450
• /
• 2012

Based upon a questionnaire survey, this research study focuses on the Japanese apparel manufacturers. The aim is to understand the characteristics that make up the marketing strategys of Japanese small to medium sized apparel manufacturers. The authors generate exploratory hypotheses, which are tested via statistical analysis of data obtained from a questionnaire survey. The hypotheses tested include, the relationships between "High Value Added" strategys and factors involved with manufacturing and marketing channels. High Value Added strategys relate to the manufacturers' competitive behaviors that create high and new value for their products. The results indicate that High Value Added strategys have positive relationships for domestic outsources, the number of outsources, use of directly operated shops and the orientation for sales in overseas markets.

• 응용통계연구
• /
• 제3권2호
• /
• pp.39-53
• /
• 1990

분균형자료의 분석과 가설검정을 위하여 표본수 $n_{ij}$가 $n_{ij}>0$인 경우와 ${n_{1j}}{/geq}0$인 경우의 인위적인 이원분류자료를 써서 여러 가지 모형하에서 주효과와 교호작용에 대한 변동을 계산하는 방법을 Searle의 R($\mid$)표기법을 이용하여 검토하였고 이들 변동과 일반적으로 널리 쓰이는 가설과의 관계를 실증적으로 재검토하였다.

• 한국정보시스템학회지:정보시스템연구
• /
• 제28권3호
• /
• pp.123-139
• /
• 2019

Purpose With the development of information and communication technologies. fintech industry is rapidly growing as a major financial service. Previous research literatures mainly focused on overall characteristics and technical aspects of fintech including security issues. Therefore, this study investigated the effect of individual features influencing the resistance of adopting a fintech service. Moreover, this paper included the guidelines in practical development of fintech mechanism. Design/methodology/approach For research purposes, this study developed research hypotheses in order to empirically examine the effect of the factors that might have a significant effect on the resistance of the wearable device. The empirical research was based on a survey which carried through 359 participants. Research hypotheses were evaluated via SPSS Statistics 21.0 and AMOS 19 statistical package program. Findings Results showed that self-efficacy and individual innovation had a significant effect on the technostress and switching cost. In addition, technostress, security and switching cost had a significant effect on the resistance of adopting a fintech service. However, more factors are needed to including in research model to generalize these results.