Go to the main menu
Skip to content
Go to bottom
REFERENCE LINKING PLATFORM OF KOREA S&T JOURNALS
> Journal Vol & Issue
Korean Journal of Applied Statistics
Journal Basic Information
Journal DOI :
The Korean Statistical Society
Editor in Chief :
Volume & Issues
Volume 9, Issue 2 - Sep 1996
Volume 9, Issue 1 - Mar 1996
Selecting the target year
Bootstrapping trimmed estimator in statistical inference
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 1~11
As an estimate of a location parameter for a given data set,
-trimmed mean has been studied for a long time by many statisticians because of its nice propoerties including robustness. However, its performance depends on the proportion of trimming say
. In this paper, we suggest a data-driven choice of
and study its validity. Also, we suggest a new estimator and consider double-bootstrap to improve its performance. By using simulation study, the proposed method is compared with the exiting one in various cases. Real data sets are also analyzed by using the proposed method.
Boostrap testing for independence in Marshall and Olkin's model under random censorship
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 13~23
In this paper, we consider the Marshall and Olkin's bivariate exponential model under random censorship for the distribution of failure times of a system with two components. We propose a bootstrap testing procedure for independence and compare the powers of it with other tests via Monte Carlo simulation.
Analysis of the 'Chukwookee' data using time series model
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 25~43
One of the main issues related to the precipitation amounts measured by the Korean raingage, Chukwookee, invented by King Sejong is the discontinuity in the time series around 1907 when the modern raingage was first used in Korea. To solve this discontinuity problem Wada(1971) reproduced the Chukwookee data but many authors questioned the validity of Wada's method. In this paper we analyze the precipitation amounts in Seoul from 1771 to 1994 using the intervention model and show that Wada's method results in the overestimation of the precipitation amounts. We also propose a reproduction method by considering monthly constant and including the rainfall of less then 2 mm and the snowfall which were ignored previously.
Reliability of essay-writing scoring in university entrance exam
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 45~53
Essay-writing, first introduced to Korean university entrance exams in 1994, is gaining its weight year by year. Especially from 1997 when Nationwide Education Reform System begins, it will be a key component of student selection criteria at Korean universities. Essay-writing's future, however, will not be that smooth unless it shows necessary validity and reliability. This study is on reliability of Essay-writing scoring, mainly from the experience of University K case. To secure solid reliability in Essay-writing scoring for the 1995 University Entrance Exam, the authors started research from the 1994 Autumn Pre-exam which was administred to potential applicants of University K following year. Total of 1,254 students took Essay-writing exam and, subsequently, their essays were graded by two professors independently. The result was not so good. The correlation between two scores was 0.27[0.54] with Cronbach alpha 0.43[0.70] for Humanity-Social Science [Natural Science-Engineering] field. So, some action for reliability improvement was inevitable. The authors considered and investigated following two alternatives. Alternative 1 [A1] : Essays are to be graded three tiems independently at the 1995 University K Entrance Exam. Scores will be given as the average of three scores. Alternative 2 [A2] : Essays are to be graded twice independently, followed by a possible third grading only if two gradings show "significant" defference. Scores are given as the third score if done or the average of first two scores otherwise.otherwise.
Biplot method algorithm and application in tire engineering
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 55~72
It is essential in modern industry that quality and procuctivity are improved continuously. To accomplish this purpose, quality control must be maintained in all parts of a company. Recently, some tire manufacture companies are beginning to show interest in quality control. They have tried to achive some results through the statistical analysis for the experimental data which has accumulated up to now and then they strive to determine the structural relationship between the design factors in tire construction and tire performance characteristics. The measurement data obtained from the construction engineering is given in multivariate form owing to the various properties found in tire design components as wll as in performance. Also it may be existed the relationship among the multimple response variables. Thus we proposes the use of the biplot graphical display as an analytic tool of data matrices with complex respects. The proposed biplots are also availalbe to understand both the underlying structure of the data and the roles played by the different components. In particular, we consider the matter of how best to use the biplots in the maltivariate analysis of variance and multiple response data. Finally we apply this method to analyze the actual data.
A Development of Multimedia Software for STatistical Education using Authoring Tool-Dice and Card Game-
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 73~82
A multimedia software for introductory education is developed based on computer simulation. Developing tools for educational software are discussed. A developed software can be used interactively in teaching of statistical basic concepts.
On estimation of the probability of Yut
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 83~94
The probability of Yut was calculated by using the physical property in previous study, but this article suggested empirical estimators for probability of Yut. In practice, physics-based probability imposes too strong assumptions, which result in the difference between the calculated probabilies and empirical relative frequencies. Experiment shows the probabilities of Yut depend on the integrated shape of Yut rather than the floor type. Maximum likelihood estimator and empirical Bayes estimators are compared and all turn out to be almost identicla for more than 40 trials. For smaller number of trials, Bayes estimators are recommended for its stability. Regression approach is also adopted as an easy-to-use method without empirical trials.
Group sequential testing methods for comparing cure rates
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 95~108
There are many clinical trials where a large portion of the patients are cured of disease. In such a case, one might be more interested in testing the differences in cure rates rather than other types of differences in failure distribution. For ethical and economic reasons, clinical trials must be repeatedly monitored for evidence of treatment benefit or harm. In this article, we examined by simulation the properties of nonparametric group sequential methods for comparing the cure rates between two treatment groups during the trial in a wide range of alternatives, censoring rates and cure rates.
More informative sequential probability ratio test
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 109~117
We introduce the more informative sequential probability ratio test(SPRT) than currently used SPRT. Though the proposed SPRT shares similar mathematical properties with the ordinary SPRT, it is less affected by the outliers and even it indicates possible existence of such outliers. Futhermore, it responds to the changes among observations more quickly than the ordinary SPRT.
Accelerated life testing data analysis using the model incorporating the random environmental effect
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 119~134
Accelerated life testing (ALT) of a system is commonly used to reduce time and cost. ALT is achieved by subjecting the test systems to more severe conditions than the normal ones to obtain estimates of life distribution under normal condition. The major interest of this research is to use a model of incorporating the common environmental effect on the components serially linked into a system-so called frailty model for the system life time distribution under each stress and to discuss the related data analysis and comparison of the model with the generally used one. The profile likelihood is used to get an initial values required to compute maximum likelihood estimates and simulation is carried for comparison.
A study on the efficiency of the multiple trial randomized response technique
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 135~143
In surveys on certain social problems which are sensitive in nature, many techniques have been introduced in order to protect evasive or untruthful answers. We suggest a multiple trial randomized response technique(MRRT) and it turns out that MRRT is feasible and more efficient by reducing the variance of the estimate than single trial RRT's investigated by Warner(1965), Mangat & Singh(1990), Mangat(1994).
Estimating using the method of adaptive searching observation
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 145~159
We propose an adaptive searching method using some spatial relations among sample points to estimate the interesting area in the spatial population. The fundamental idea is to observe the neighboring sample points when a sample point is satified with some condition of an adaptive searching observation. For obseving the sample points with this method to estimate the area the sample size is decreased. From this result, we may expect to reduce the cost and time consuming in observation the sample points and to draw the shape of the interesting area without prior information of an spatial population. Some analytical simulation results are also presented.
A study for the efficiency of the cut-off method in highly skewed populations
Korean Journal of Applied Statistics, volume 9, issue 2, 1996, Pages 161~169
In the design of the sampling, it is important to make a decision about the size of the sample to be selected from the population. We often have a problem to get the optical size of the sample to be considered for cost and time expended for selecting sample unit from highly skewed population. In this case, we give a graphical criterion with Take-all Stratum rate to choose a method and also illustrate the efficiency between the Neyman allocation and the cut-off method with real data.