• Journal of the Korean Data and Information Science Society
• /
• 제28권5호
• /
• pp.1055-1067
• /
• 2017

지금까지 연구되어진 한국의 장애인 우울에 미치는 요인들을 찾는 연구들은 조사된 장애인별로 반복 측정된 자료임에도 대부분 그들 자료들이 독립이라는 가정 하에 통계적 분석을 하였다. 본 연구에서는 3년마다 반복 측정된 한국복지패널의 장애인 관련 부가조사 자료를 이용하여 장애인 우울에 미치는 요인들이 무엇인지를 반복 측정의 효과를 포함하는 선형혼합모형으로 분석한다. 각 장애인별로 얻어진 반복 측정 자료들의 종속성에 적절한 상관행렬을 고려하고, 회귀계수인 고정효과 뿐만 아니라 각 장애인별 특성을 반영하는 임의효과를 포함시켜 장애인 우울에 유의한 변수들을 찾고 최종모형을 구한다. 고정효과만 고려한 모형으로부터 얻어진 잔차들은 각 장애인별 잔차들의 평균이 0 근처에 있지 않다는 것을 보여주고, 임의효과를 포함하고 있는 최종모형의 잔차들의 잔차도와 분위수대조도를 제시하여 최종모형은 각 장애인별 잔차들의 평균의 문제점을 해결한다는 것을 보여주고자 한다.

• 한국시뮬레이션학회논문지
• /
• 제10권3호
• /
• pp.31-46
• /
• 2001

Repeated measurements on units under different conditions are common in biological and biomedical studies. In a number of growth and pharmacokinetic studies, the relationship between the response and the covariates is assumed to be nonlinear in some unknown parameters and the form remains the same for all units. Nonlinear random coefficient models are used to analyze such repeated measurement data. Extended least squares methods are proposed in the literature for estimating the parameters of the model. However, neither objective function has closed form expression in practice. This paper proposes Monte Carlo methods to estimate the objective functions and the corresponding estimators. A simulation study that compare various methods is included.

• 한국의류학회지
• /
• 제40권4호
• /
• pp.641-649
• /
• 2016

Highly precision body measurements represent basic data required by industry and researches who wish to utilize information about the human body. The proficiency and expertise of the measurers have a significant influence on the error and accuracy of data when various parts from multiple subjects' bodies are measured. Therefore, in order to measure accurate body measurements (when measuring bodies directly), it is necessary to conduct objective analyses on errors. This study calculated the Relative Technical Error of Measurement (%TEM) using data that measured each of 24 subjects and discussed errors and methods to reduce errors by conducting comparison analysis based on measured items and objects. The result of analysis indicated that the errors based on age and gender of the objects of measurement were minor; however, there were comparatively distinct differences in measured errors based on measured items. 'Right and left Shoulder Angle' for all measured subjects displayed the greatest errors and standard deviations. 'Height' dimension, Lateral Malleolus Height and Head Height had big errors; in addition, 'Circumference', Neck Base Circumference and Armscye Circumference also had big errors. More careful measurements of such items with big errors require additional educational plan such as a proposal for more objective and detailed measurement methods. Items with small errors but big standard deviations such as Waist Circumference, Calf Circumference, Minimum Leg Circumference, Chest Circumference, Hip Circumference and Waist Circumference confirmed that errors for them greatly decreased with repeated experiments and resultant measurers increased proficiency; consequently, repeated measuring experiments for these items greatly enhance accuracy.

• 한국경영과학회지
• /
• 제26권1호
• /
• pp.109-116
• /
• 2001

The detection of corporate failures is a subject that has been particularly amenable to cross-sectional financial ratio analysis. In most of firms, however, the financial data are available over past years. Because of this, a model utilizing these longitudinal data could provide useful information on the prediction of bankruptcy. To correctly reflect the longitudinal and firm-specific data, the generalized linear model with assuming the first order AR(autoregressive) process is proposed. The method is motivated by the clinical research that several characteristics are measured repeatedly from individual over the time. The model is compared with several other predictive models to evaluate the performance. By using the financial data from manufacturing corporations in the Korea Stock Exchange (KSE) list, we will discuss some experiences learned from the procedure of sampling scheme, variable transformation, imputation, variable selection, and model evaluation. Finally, implications of the model with repeated measurement and future direction of research will be discussed.

• 대한치과기공학회지
• /
• 제41권3호
• /
• pp.211-217
• /
• 2019

Purpose: The purpose of this study is to evaluate the repeated measurement stability of scans related to dentition type. Methods: A normal model and the crowding and diastema models are also duplicated using duplicating silicon. After that, a plaster model is made using a plaster-type plaster on the duplicate mold, and each model is scanned 5 times by using an extraoral scanner. The gingival part and molar part were deleted from the 3D STL file data obtained through scanning. Using the 3D stl file obtained in this way, data is nested between model groups. Thereafter, RMS values obtained were compared and evaluated. The normality test of the data was performed for the statistical application of repeated measurements with dentition type, and the normality was satisfied. Therefore, the one-way ANOVA test, which is a parametric statistical method, was applied, and post-tests were processed by the Scheffe method. Results: The average size of each RMS in the Normal, Diastema, and Crowding groups was Normal> Crowding> Diastema. However, the standard deviation was in the order of Crowding> Normal> Diastema. The average value of each data is as follows. Diastema model was the smallest ($5.51{\pm}0.55{\mu}m$), followed by the crowding model ($12.30{\pm}2.50{\mu}m$). The normal model showed the maximum error ($13.23{\pm}1.06{\mu}m$). Conclusion: There was a statistically significant difference in the repeatability of the scanning measurements according to the dentition type. Therefore, you should be more careful when scanning the normal intense or crowded dentition than scanning the interdental lining. However, this error value was within the range of applicable errors for all clinical cases.

• 재활간호학회지
• /
• 제17권1호
• /
• pp.48-55
• /
• 2014

Purpose: In this study, we attempted to provide an evidence for the effects of meridian acupressure on relieving and preventing constipation in the institutionalized elderly. Methods: The research design was a non-equivalent control group, non-synchronized design. The subjects consisted of 31 institutionalized elderly (experimental group: 16, control group: 15). The experimental group was given meridian acupressure for 10 minutes daily for 2 weeks. The data was analyzed by the ${\chi}^2$-test, Fisher's exact test, t-test and repeated measures ANOVA. Results: The number of bowel movements per week of the experimental and control group performed meridian acupressure verified by repeated-measures analysis of variance revealed that interaction existed between the meridian acupressure availability and the measurement point (F=98.183, p<.001). Repeated measures analysis of variance to compare the changes in Bristol stool form scale scores of the experimental and control group performed meridian acupressure revealed that interaction existed between the meridian acupressure availability and the measurement point (F=48.896, p<.001). Conclusion: The results of this study show the meridian acupressure is a useful nursing intervention on constipation in the institutionalized elderly.

• 응용통계연구
• /
• 제28권2호
• /
• pp.361-369
• /
• 2015

연속측도의 반응변수가 반복측정된 실험 자료의 분석을 위해 흔히 선형혼합모형이 사용된다. 그러나, 잔차의 분포가 이분산성이거나 비정규성을 가질 때 표준적인 선형혼합모형은 적절하지 않은 결과를 가져온다. 잔차의 분포가 두터운 꼬리를 가진 비정규분포를 보이는 타이어 필드시험 데이터를 로버스트 선형혼합모형에 적합시킴으로써 보다 더 정확하고 신뢰할 수 있는 분석결과를 얻을 수 있다. 추가적으로 신뢰성 분석 결과를 제시한다.

• The Korean Journal of Pain
• /
• 제30권4호
• /
• pp.243-249
• /
• 2017

Pain is subjective, while statistics related to pain research are objective. This review was written to help researchers involved in pain research make statistical decisions. The main issues are related with the level of scales that are often used in pain research, the choice of statistical methods between parametric or nonparametric statistics, and problems which arise from repeated measurements. In the field of pain research, parametric statistics used to be applied in an erroneous way. This is closely related with the scales of data and repeated measurements. The level of scales includes nominal, ordinal, interval, and ratio scales. The level of scales affects the choice of statistics between parametric or non-parametric methods. In the field of pain research, the most frequently used pain assessment scale is the ordinal scale, which would include the visual analogue scale (VAS). There used to be another view, however, which considered the VAS to be an interval or ratio scale, so that the usage of parametric statistics would be accepted practically in some cases. Repeated measurements of the same subjects always complicates statistics. It means that measurements inevitably have correlations between each other, and would preclude the application of one-way ANOVA in which independence between the measurements is necessary. Repeated measures of ANOVA (RMANOVA), however, would permit the comparison between the correlated measurements as long as the condition of sphericity assumption is satisfied. Conclusively, parametric statistical methods should be used only when the assumptions of parametric statistics, such as normality and sphericity, are established.

• 대한인간공학회지
• /
• 제14권2호
• /
• pp.75-85
• /
• 1995

The purpose of this study is to develop a force-platform system suitable for real situations in industry. The developed system is a 3-axis measurement type. It consists of force-transmission, force measuring circuit, and the software that analyzes the measured force. The force-transmission transmits forces to 3 axes, which has a physical structure to minimize the interference among those axes. Force measuring circuit consists of DSP (digital signal processor) for flexible disposal of change of measurement algorithm, elimination of noise and maintenance of precision. The functions of the software are the calibration which revises the measurement error occuring during data acquisition, and various analyses of forces. The result of the experiment shows that the developed system has about 1% measurement error, is stable for repeated experiments, and is not effected by temperature change.

• 한국초등과학교육학회지:초등과학교육
• /
• 제28권3호
• /
• pp.263-276
• /
• 2009

The purpose of this study was to analyze the elementary students' selection of representable value and confident method that appear in measuring activities by using a microgenetic method. The participants were seven elementary students in the fourth grade. They performed the same measuring activities six times for the study period. Data were collected by interview and observation with their activity recording papers and video tape transcription. Their activities were recorded and documented for the analysis. Results were as follows. First, in the time measuring activity, elementary students developed desirably as their measuring experience increased, for example they selected a representable value in use of a repeated measurement and used a various method in the domain of a time measurement and they showed an increase of a quantitative observation in the volume domain except in the length domain. Second, in a confident method of a representable value, though they must rely upon a repeated measurement, they only measure repeatedly in the time domain. Also in the time domain, it doesn't get accomplished a exact confidence of a representable value at a shortage of skill about a measurement. Accordingly this study will be implications for teachers to teach a handling abilities of measuring instruments to elementary students and to be promote understanding a nature of measurement.