Implementation of Estimation and Inference on the Web

Kang, Heemo,Sim, Songyong 한국통계학회 2000 Communications for statistical applications and me Vol.7 No.3

An electronic statistics text on the web is implemented. The introduced text provide interactive instructions on the statistical estimation and inference. As a by-product, we also provide a calculation of quantiles and p-value of t-distribution and standard normal distribution. This program was written in JAVA programming language.

정규 모집단의 모평균 차이 검정에서 표본크기와 검정력 계산의 구현

심송용,최규혁,Sim, Songyong,Choi, Kyuhyeok 한국데이터정보과학회 2013 한국데이터정보과학회지 Vol.24 No.3

임상시험 등 다양한 분야를 중심으로 제1종 오류확률과 검정력을 함께 고려하여 표본크기를 결정하는 경우가 늘어나고 있다. 이런 경향은 표본을 많이 얻을 수 없는 연구에서 더욱 뚜렷하다. 본 연구에서는 독립인 두 개의 정규모집단에서 두 그룹의 분산과 표본수가 같지 않을 때의 모평균 차이에 대한 검정에서 제1종 오류와 제2종 오류를 모두 고려한 경우 두 그룹의 필요한 표본크기를 결정하는 과정을 살펴보고 이를 웹사이트를 통해 구현하였다. 또한 주어진 표본크기와 유의수준에 의한 검정력 계산도 함께 구현하였다. 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.

Dietary Habits Are Associated With School Performance in Adolescents

Kim, So Young,Sim, Songyong,Park, Bumjung,Kong, Il Gyu,Kim, Jin-Hwan,Choi, Hyo Geun Williams & Wilkins Co 2016 Medicine Vol.95 No.12

<P><B>Abstract</B></P><P>Several studies suggest that dietary habits are associated with poor academic performance. However, few studies have evaluated these relations after adjusting for numerous confounding factors. This study evaluated the frequency of various diet items (fruit, soft drinks, fast foods, instant noodles, confections, vegetables, and milk) and the regularity of meal times (breakfast, lunch, and dinner) all at once.</P><P>A total of 359,264 participants aged from 12 to 18 years old were pooled from the Korea Youth Risk Behavior Web-based Survey (KYRBWS) for the 2009 to 2013 period. Dietary habits over the last 7 days were surveyed, including the regularity of consuming breakfast, lunch and dinner and the frequency of eating fruits, soft drinks, fast foods, instant noodles, confections, vegetables, and milk. Physical activity, obesity, region of residence, subjective assessment of health, stress level, economic level, and parental education level were collected from all of the study participants. School performance was classified into 5 levels. The adjusted odds ratios (AORs) of dietary habits for school performance were analyzed using multinomial logistic regression analyses with complex sampling. Structural equation modeling was used to analyze the effects of diet factors on school performance while considering the effects of other variables on both diet factors and school performance.</P><P>Frequent intakes of breakfast (AOR = 2.34, 95% confidence interval [CI] = 2.20–2.48), fruits (AOR = 1.73, 95% CI = 1.62–1.86), vegetables (AOR = 1.48, 95% CI = 1.37–1.61), and milk (AOR = 1.35, 95% CI = 1.28–1.43) were related to high levels of school performance (each with <I>P</I> < 0.001). In contrast, soft drinks (AOR = 0.42, 95% CI = 0.38–0.46), instant noodles (AOR = 0.62, 95% CI = 0.55–0.70), fast food (AOR = 0.83, 95% CI = 0.72–0.96), and confectionary (AOR = 0.86, 95% CI = 0.80–0.93) were negatively associated with school performance (each with <I>P</I> < 0.001).</P><P>This study confirms previous studies of school performance and dietary habits that find a positive association with eating breakfast and consuming fruits and milk and a negative relation with soft drinks, instant noodles, fast foods, and confections.</P>

High Incidence of Bell’s Palsy After Mastoidectomy: A Longitudinal Follow-up Study

Choi, Hyo Geun,Sim, Songyong,Hong, Sung Kwang,Park, Su-kyoung,Lee, Hyo-Jeong,Chang, Jiwon Lippincott Williams Wilkins 2017 Otology & Neurotology Vol.38 No.10

OBJECTIVE:: The objective of this study was to compare the prevalence of Bell’s palsy in participants who underwent mastoidectomy (to treat chronic otitis media) and nonmastoidectomy participants (control). METHODS:: Using the national cohort study from the Korean Health Insurance Review and Assessment Service, mastoidectomy patients (2,045) and control participants (8,180) were matched 1:4 for age, sex, income, and region of residence. The prevalence of Bell’s palsy in both the groups was measured from 0 to 10 years postoperation. RESULTS:: In a sample of 1,025,340 Korean individuals, 7,070 were diagnosed or treated with Bell’s palsy between 2002 and 2013; the annual incidence of Bell’s palsy was 0.057%. The overall prevalence of Bell’s palsy was three times higher in the mastoidectomy group (1.27%) than control group (0.49%) (p < 0.001). The prevalence of Bell’s palsy was different between the two groups in postoperative 0 year: 0.78% for the mastoidectomy group versus 0.01% for the control group (p < 0.001). CONCLUSION:: Although we could not verify the laterality, the prevalence of Bell’s palsy was increased in chronic otitis media patients treated with mastoidectomy patients compared with controls, especially within a year after surgery.

A Study on Data Mining Using the Spline Basis

Lee, Sun-Geune,Sim, Songyong,Koo, Ja-Yong 한국통계학회 2004 Communications for statistical applications and me Vol.11 No.2

Due to a computerized data processing, there are many cases when we encounter a huge data set. On the other hand, advances in computing technologies make it possible to deal with a huge data set. One important area is the data mining. In this paper we consider data mining when the dependent variable is binary. The proposed method is to use the poly-class model when the independent variables consists of continuous and discrete variables. An example is provided.

Regression and Correlation Analysis via Dynamic Graphs

Kang, Hee Mo,Sim, Songyong 한국통계학회 2003 Communications for statistical applications and me Vol.10 No.3

In this article, we propose a regression and correlation analysis via dynamic graphs and implement them in Java Web Start. For the polynomial relations between dependent and independent variables, dynamic graphics are implemented for both polynomial regression and spline estimates for an instant model selection. The results include basic statistics. They are available both as a web-based service and an application.

Development of a Tool to Measure the Need for Child Hospice Care in Families of Children with Cancer

Kyung-Ah Kang,Songyong Sim,Shin-Jeong Kim 한국간호과학회 2005 한국간호과학회 학술대회 Vol.2005 No.10

브라운다리 근사를 통한 확산모형의 우도 근사법

이은경,심송용,이윤동,Lee, Eun-kyung,Sim, Songyong,Lee, Yoon Dong 한국통계학회 2015 응용통계연구 Vol.28 No.5

확산모형은 입자의 운동현상과 금융자산의 미시적 가격변동을 설명하기 위하여 사용되는 수리적 모형이다. 확산모형의 추정방법에 관한 논의는 다양한 분야에서 이루어져 왔다. 통계학적 관점에서 우도적 방법에 기반한 확산모형의 추정방법을 개발하려는 시도가 계속되어 왔다. 이산시간 간격으로 관측된 자료를 이용하여 확산모형을 추정할 때 최대우도 추정법을 적용하기 위해서는 확산모형에 대한 전이확률 밀도함수를 구해야 한다. 본 연구에서는 확산모형의 전이확률밀도를 근사하기 위하여, 정규분포를 따르는 확률변수를 이용하여 브라운다리 확률과정에 대한 경로적분을 대체하는 방법을 제안하고, 그 수치적 성질을 다른 방법들과 비교한다. Diffusion is a mathematical tool to explain the fluctuation of financial assets and the movement of particles in a micro time scale. There are ongoing statistical trials to develop an estimation method for diffusion models based on likelihood. When we estimate diffusion models by applying the maximum likelihood estimation method on data observed at discrete time points, we need to know the transition density of the diffusion. In order to approximate the transition densities of diffusion models, we suggests the method to approximate the path integral of the random process with normal random variables, and compare the numerical properties of the method with other approximation methods.

요인적재값 가중치를 사용한 평가 시스템에 대한 연구

이기원,심송용,Lee, Kee-Won,Sim, Songyong 한국데이터정보과학회 2016 한국데이터정보과학회지 Vol.27 No.5

추상적 개념을 계량화 하기 위해 상대적으로 구체적인 여러 개의 문항을 조사한 후 이들 점수의 합 또는 이들 점수를 표준화한 후 합을 구하는 리커드 (Likert) 척도 (합산등급척도법)를 많이 사용한다. 합산등급척도법은 각 항목의 크기가 차이가 많이 나는 경우에 원자료가 아닌 표준화 값을 사용하여 합하기도 한다. 이와 같은 상황은 평가 시스템에서도 발생한다. 예를 들어 기초지방자치단체들을 발전정도에 따라 분류하기 위해 인구, 세수현황 등의 값을 표준화하고 이를 단순합산하여 분류의 기초로 사용할 수 있다. 본 연구에서는 위의 같은 추상적 개념의 수치화 또는 평가 시스템에 많이 적용되는 합산등급척도법의 문제점을 개선하는 한 방법으로 가중치를 자료에서 계산하는 데이터 구동 방식의 평가 시스템을 제안하고, 이 시스템을 실자료에 적용한다. To quantify an concept we often use Likert summated rating scale of original or standardized variables in case the variables are relatively less. When variables have different scales, standardized values tends to be used rather than the original values. This is also true in evaluating systems. For example, we may use standardized values of local tax levy, population, and etc. and use the summed value of the standardized values to access the degrees of development. In this paper, we propose using a data-driven weighted sum for a scoring system and the way how to obtain the weights. We apply the proposed method to a real data set and find that proposed method is better than the usual summated rating scale.

Excessive Sleep and Lack of Sleep Are Associated With Slips and Falls in the Adult Korean Population : A Population-Based Cross-Sectional Study

Kim, So Young,Kim, Sung-Gyun,Sim, Songyong,Park, Bumjung,Choi, Hyo Geun Williams & Wilkins Co 2016 Medicine Vol.95 No.4

<▼1><P>Supplemental Digital Content is available in the text</P></▼1><▼2><P><B>Abstract</B></P><P>Few studies have evaluated the impacts of excessive sleep duration on falls. This study investigated the associations between sleep duration and falls among Korean adults in a wide range of age groups while adjusting for numerous confounding factors. Data collected from study participants ranging in age from 19 to 109 years old were analyzed from the 2013 Korean Community Health Survey (KCHS). Sleep duration was divided into 5 groups: ≤5, 6, 7, 8, and ≥9 hours per day. The relations between sleep duration and falls (≥1 time or ≥2 times per year) were analyzed using simple and multiple logistic regression analyses with complex sampling. Age, sex, days of vigorous or moderate physical activity, income, education, alcohol use, smoking, stress, obesity, hypertension, diabetes mellitus, hyperlipidemia, stroke, angina or myocardial infarction, arthritis, and asthma were controlled for as confounding factors. Associations between sleep duration and falls were analyzed in 19 to 40, 41 to 60, and 61+ year age groups. Furthermore, the relations between sleep duration and indoor versus outdoor falls were analyzed. Both ≤6 and ≥8 hours of sleep per day were significantly associated with an increased incidence of falls (≥1 time and ≥2 times per year) in the overall adult population (<I>P</I> < 0.001 in both instances). In a subgroup analysis, sleep durations of ≤5 and ≥9 hours were significantly associated with an increased incidence of falls (≥1 time a year) in each age group. Six hours of sleep was not significantly associated with falls (≥2 times per year) in the 61+ year age group, and 8 and 9 hours of sleep were not significantly associated with falls (≥2 times a year) in the 19 to 40 year age group. This study demonstrated that long as well as short sleep durations are associated with an increased incidence of falls. However, these relations were not evident in elderly populations with short sleep durations or in young adults with long sleep durations.</P></▼2>