• Journal of the Korean Society of Hazard Mitigation
• /
• v.7 no.5
• /
• pp.79-88
• /
• 2007

This study evaluated the statistical stationarity of rainfall quantiles as well as the rainfall itself. The conventional methodologies like the Cox-Stuart test for trend and Dickey-Fuller test for a unit root used for testing the stationarity of a time series were applied and evaluated their application to the rainfall quantiles. As results, first, no obvious increasing or decreasing trend was found for the rainfall in Seoul, which was also found to be a stationary time series based on the Dickey-Fuller test. However, the Cox-Stuart test for the rainfall quantiles show some trends but not in consistent ways of increasing or decreasing. Also, the Dickey-Fuller test for a unit root shows that the rainfall quantiles are non-stationary. This result is mainly due to the difference between the rainfall data and rainfall quantiles. That is, the rainfall is a random variable without any trend or non-stationarity. On the other hand, the rainfall quantiles are estimated by considering all the data to result in high correlation between their consecutive estimates. That is, as the rainfall quantiles are estimated by adding a stationary rainfall data continuously, it becomes possible for their consecutive estimates to become highly correlated. Thus, it is natural for the rainfall quantiles to be decided non-stationary if considering the methodology used in this study.

• The Korean Journal of Applied Statistics
• /
• v.23 no.1
• /
• pp.123-138
• /
• 2010

After selecting a group of women with premenstrual syndrome based on daily distress scores of 28 days, one needs to estimate threshold for the change of symptoms, which would be useful for the clinician's diagnosis in hospitals. However, a test of whether a change has occurred has to precede the estimation of the threshold. In this paper, we apply parametric and nonparametric testing methods to an example data obtained from a group of women. Nonparametric method does not assume any distributional form of distress scores and parametric testing method is based on the normal distributions of linear regression lines. Therefore, the optimal situation of both methods would be different and we will assess it with a simulation study.

• The Korean Journal of Applied Statistics
• /
• v.30 no.4
• /
• pp.503-512
• /
• 2017

In test with multiple choices, it is necessary to have the position of correct answers of each question spreaded evenly over all the questions in order to minimize the influence of answering tendency of test takers with preference to specific position of multiple choices. The scores of tests with correct answers in specific positions would not reflect exactly the academic aptitude of examinees who do not know correct answers but have the biased answering tendency. In this paper, we have randomness test for the positioning of correct answers at the 2017 College Scholastic Ability Test (CSAT) using Bartels rank test, the Wald-Wolfowitz runs test, the turning point test, the Cox Stuart trend test, the difference sign test and the Mann-Kendall tank test, etc. We also do independence test between the location of correct answer and the allocation of score in each question, for it may result in overestimating the test-taker with specific position preference in marking correct answers.