• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.262-264
• /
• 2005

We propose a filter design method to suppress the effect of gyroscope mixed random errors at INS system level. It is based on the result that mixed random errors can be represented by a single equivalent ARMA model. At first step, the time difference of equivalent ARMA process is performed, which consider the characteristic of indirect feedback Kalman filter used in INS filter. Next, a state space conversion of time differenced ARMA model is achieved. If the order of AR is greater than that of MA, the controllable or observable canonical form is used. Otherwise, we introduce the state equation of which the state variable is composed of the ARMA model output and several step ahead predicts of that. At final step, a complete form state equation is presented. The simulation results shows that the proposed method gives less transient error and better convergence compared to the conventional filter which assume the mixed random errors as white noise.

• 한국항공우주학회지
• /
• 제34권4호
• /
• pp.47-52
• /
• 2006

불규칙 혼합잡음의 등가 ARMA 모델 표현을 사용하여 자이로의 불규칙 혼합잡음을 고려하는 보조항법시스템 칼만필터 설계 방법을 제안한다. 필터 설계 절차는 먼저 보조항법 시스템에 사용되는 필터는 간접 되먹임 칼만필터임을 고려하여 등가 ARMA 모델로 표현된 자이로 불규칙 잡음의 시간 차분을 구한다. 다음으로 시간 차분된 ARMA 모델을 상태 방정식으로 표현하는데 AR과 MA 차수에 따라 두 가지로 나누어진다. 먼저 AR 차수가 큰 경우 가제어 혹은 가관측 특이형태를 사용한다. MA 차수가 큰 경우에는 몇 단계 이후의 예측치를 상태변수로 하는 상태방정식을 사용하는데, 이때 자이로 출력을 보상하는 값에 따라 다시 고차수 필터와 저차수 필터로 구분된다. 마지막으로 자이로 불규칙 잡음을 보조항법시스템 칼만필터에 포함시켜 최종적인 필터 모델을 얻는다. 시뮬레이션 결과를 통하여 제안된 고차수 및 저차수 필터 모두 혼합잡음을 백색잡음으로 간주한 기존의 필터보다 항법오차를 감소시킬 수 있음을 보임으로써 그 효용성을 제시한다.

• Journal of the Korean Statistical Society
• /
• 제28권2호
• /
• pp.225-234
• /
• 1999

We consider the following semi-parametric non-linear mixed effect regression model : y\ulcorner=f($\chi$\ulcorner;$\beta$)+$\sigma$$\mu$($\chi$\ulcorner)+$\sigma$$\varepsilon$\ulcorner,i=1,…,n,y*=f($\chi$;$\beta$)+$\sigma$$\mu$($\chi$) where y'=(y\ulcorner,…,y\ulcorner) is a vector of n observations, y* is an unobserved new random variable of interest, f($\chi$;$\beta$) represents fixed effect of known functional form containing unknown parameter vector $\beta$\ulcorner=($\beta$$_1$,…,$\beta$\ulcorner), $\mu$($\chi$) is a random function of mean zero and the known covariance function r(.,.), $\varepsilon$'=($\varepsilon$$_1$,…,$\varepsilon$\ulcorner) is the set of uncorrelated measurement errors with zero mean and unit variance and $\sigma$ is an unknown dispersion(scale) parameter. On the basis of finite-sample, small-dispersion asymptotic framework, we derive an absolute lower bound for the asymptotic mean squared errors of prediction(AMSEP) of the regular-consistent non-linear predictors of the new random variable of interest y*. Then we construct an optimal predictor of y* which attains the lower bound irrespective of types of distributions of random effect $\mu$(.) and measurement errors $\varepsilon$.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.69.2-69
• /
• 2002

Process measurements contain random and gross errors and the size estimation of gross errors is required for production accounting. Mixed integer programming technique had been applied to identify and estimate the gross errors simultaneously. However, the compensate model based on mixed integer programming used all measured variables or spanning tree as gross error candidates. This makes gross error estimation problem combinatorial or computationally expensive. Mixed integer programming with test statistics is proposed for computationally inexpensive gross error identification /estimation. The gross error candidates are identified by measurement test and the set of gross error candidates are...

• 대한수학회논문집
• /
• 제28권2호
• /
• pp.243-267
• /
• 2013

This paper is dedicated to study a new class of general nonlinear random A-maximal $m$-relaxed ${\eta}$-accretive (so called (A, ${\eta}$)-accretive [49]) equations with random relaxed cocoercive mappings and random fuzzy mappings in $q$-uniformly smooth Banach spaces. By utilizing the resolvent operator technique for A-maximal $m$-relaxed ${\eta}$-accretive mappings due to Lan et al. and Chang's lemma [13], some new iterative algorithms with mixed errors for finding the approximate solutions of the aforesaid class of nonlinear random equations are constructed. The convergence analysis of the proposed iterative algorithms under some suitable conditions are also studied.

• 대한안전경영과학회:학술대회논문집
• /
• 대한안전경영과학회 2011년도 춘계학술대회
• /
• pp.701-705
• /
• 2011

The paper reviews Cronbaha's coefficient to measure a single source of error. On the contrary to classical measurement theory, the generalizability study can be used in the social survey design to calculate Generalizability Index (GI) and Dependability Index (DI) for measuring multiple sources of errors of behavior evaluation. The study proposes application guidelines to implement R:($A{\times}B$) mixed models that are composed of random factor and fixed factor.

• 응용통계연구
• /
• 제34권5호
• /
• pp.697-710
• /
• 2021

상대오차를 이용한 예측법은 상대오차(혹은 퍼센트오차)가 중요시되는 분야, 특히 계량경제학이나 소프트웨어 엔지니어링, 또는 정부기관 공식통계 부분에서 기존 예측방법 외에 선호되는 예측방법이다. 그 동안 상대오차를 이용한 예측법은 선형 혹은 비선형 회귀분석 뿐 아니라, 커널회귀를 이용한 비모수 회귀모형, 그리고 정상시계열분석에 이르기까지 그 범위가 확장되어 왔다. 그러나, 지금까지의 분석은 고정효과(fixed effect)만을 고려한 것이어서 임의효과(random effect)에 관한 상대오차 예측법에 대한 확장이 필요하였다. 본 논문의 목적은 상대오차예측법을 일반화선형혼합모형(GLMM)에 속한 감마회귀(gamma regression), 로그정규회귀(lognormal regression), 그리고 역가우스회귀(inverse gaussian regression)의 패널자료(panel data)에 적용시키는데 있다. 이를 위해 실제 자동차 보험회사의 손해액 자료를 사용하였고, 최량예측량과 최량상대오차예측량을 각각 적용-비교해 보았다.

• 응용통계연구
• /
• 제21권6호
• /
• pp.923-932
• /
• 2008

우리는 자명하지 않은 상관 구조를 갖는 복잡한 다변량 자료에 직면하는 경우가 있다. 예를 들어 군집 구조 자료의 경우 생략된 변수들이 한 개 이상의 관측값에 동시적으로 영향을 줄 수 있기 때문에 결과들 간에 상관 구조를 모형화하는 것은 추정량의 효율성과 정확한 표준오차의 계산 등의 타당한 추론을 위해서 중요하다 관측값들 간에 종속성을 두는 표준 방법으로는 관측 값들이 관찰되지 않은 어떤 변수를 공유한다고 가정하는 것인데, 이러한 가정에 대해 본 연구에서는 다수준 모형을 고려한 상관된 임의효과 모형을 적합시켰다. 추정은 준모수적 접근방법으로 임의계수 분포에 대한 모수적 가정 없이 유한혼합 EM-알고리즘을 통하여 수행되었다.

• 가정과삶의질연구
• /
• 제33권2호
• /
• pp.35-51
• /
• 2015

The purpose of this study was to provide information for research about residents' opinion toward the physical environments of elderly facilities, through the analysis and investigation on the research methodology of foreign academic journal articles from 1990 to 2014. The study results were as follows: Firstly, purposive sampling was a large majority of both facilities and elderly residents. In quantitative studies, many researchers have conducted simple random, cluster, or stratified sampling. Diverse facilities in area, size, location, and etc. should be considered for participation. The qualifications for residents' participation should be considered as well, so that they all could have autonomy for study participation. Secondly, questionnaire and semi-structured guide were likely to be used in independent and resident care facilities. On the other hand in assisted living and long-term care facilities, open questions and visual material were used as well. A compatible scale should be developed so that elderly having variable functional level could participate independently in the study. Thirdly, in data collection process, compliance with research ethics and well trained interviewer's skill were important for residents' active responses and minimization of response errors. Enough research period of time and mixed study in data collection will decrease the response error.

• 대한수의학회지
• /
• 제38권3호
• /
• pp.544-552
• /
• 1998

A random study of 574 dairy farms in Gyeongnam area was designed to determine 1) management factors that may be associated with the occurrence of drug residues; 2) the dairy farmer's attitudes and knowledge about residues; 3) how these variables influence the occurrence of residues in dairy cattle. Management factors perceived as having the greatest influence on drug residues in milk were insufficient knowledge about withdrawal periods, errors due to hired help, insufficient identification and record of animals treated for mastitis, dry cow treatment for mastitis, and metritis treatment. Seventy-one percent of farms with residues problem used mixed own feeds compared with twenty-nine percent of farms with premedicated feeds. Factors significantly associated with the occurrence of residues were herd size, increased number of hired persons, increased frequency of use of mixed-own feeds, category of medicated feed, and producer's attitude toward the public health significance of residues. Our findings suggest that residue occurrence was mainly associated with errors due to hired help, insufficient knowledge about withdrawal periods, poor animal identification and records of treatment animals and use of medicated feeds. Any residue avodiance educational program needs to stress how to deal with these factors. This educational program should be directed to dairy farmers and employees, especially temporary employees. In addition, Dairy farmer's attitudes and knowledge about drug residues need to be improved. More evidence on the public healths significance of residues should be available to them. Because belife in importance of public healths concerns was related to successful residue avidance and because 81.3% of the dairy farmers with residue problem thought public healths concerns were less important than economic ones, it would be helpful to provide educational programs specifically directed to this issue. it may be useful to provide programs not only for the dairy farmers but also change of their concerns about on the public healths.