• 제목/요약/키워드: Mixed-effect

검색결과 4,822건 처리시간 0.036초

불균형 자료에서 AIC를 이용한 선형혼합모형 선택법의 효율에 대한 모의실험 연구 (Simulation Study on Model Selection Based on AIC under Unbalanced Design in Linear Mixed Effect Models)

  • 이용희
    • 응용통계연구
    • /
    • 제23권6호
    • /
    • pp.1169-1178
    • /
    • 2010
  • 본 논문은 불균형 자료에서 선형혼합모형에 적용되는 Akaike Information Criterion(AIC)의 효율에 대한 연구이다. Vaida와 Balanchard (2005)에 의해 제안된 cAIC(conditional AIC)는 mAIC(marginal AIC)가 임의효과의 예측에 대한 불확실성을 모형선택에서 반영하지 못하는 단점을 극복할 수 있는 방법이다. cAIC에 대한 이론적인 성질과 확장은 Liang 등 (2008)과 Greven과 Kneib (2010)에 의하여 연구되었다. cAIC의 형태는 자료의 구조에 영향을 받지는 않지만 선형혼합모형에서 모수의 추정 효율은 자료의 불균형의 정도에 따라 많은 영향을 받는 것이 알려져 있다. 기존의 연구에서 실시한 모든 모의실험이 자료가 균형인 경우에만 실행되어 자료의 불균형이 AIC에 근거한 혼합모형 선택 방법의 효율에 어떤 영향을 미치는지 알려져 있지 않다. 본 논문은 자료의 불균형이 모형선택 방법의 효율에 미치는 영향을 모의실험을 통하여 알아보았다. 자료의 불균형이 심해짐에 따라 AIC에 근거한 모형선택방법은 복잡한 모형을 선택하는 경향이 낮아짐을 보였다.

혼합폐지의 탈묵효율에 미치는 지방산과 비이온성 계면활성제의 영향 (Effect of Fatty Acid and Non-ionic Surfactant on the Deinkability of Mixed Recovered Paper)

  • 서진호;최도침;류정용;정성현
    • 펄프종이기술
    • /
    • 제47권2호
    • /
    • pp.34-41
    • /
    • 2015
  • Recovered paper has been widely used as a main raw material of papermaking in Korea. Recycling of recovered paper helps to reduce production cost and preserve an environment. To recycle recovered paper efficiently, de-inking is a key process in recycling mills. De-inking process would be affected by various influencing factors such as the type of de-inking agent, mixed ratio of recovered paper, season, and process conditions. In this study, fatty acid and nonionic surfactant were used as the de-inking agent in froth-flotation process of mixed recovered paper. De-inking properties of mixed recovered paper were investigated according to the addition level of each chemical. Nonionic surfactant had a small effect on de-inking efficiency of mixed recovered paper due to decreased reject. As the additional level of fatty acid increases, fragmented ink particles increased and then optical properties of recycled paper decreased because fragmented ink particles adsorbed onto the fiber surface.

Endophyte 감염과 무감염 종자의 혼파 비율이 톨 페스큐의 endophyte 감염율 , 수량 및 사료가치에 미치는 영향 (Effect of the Mixed Rates of Endophyte-free and -infected Seen on the Endophyte Frequency , Dry Matter Yield and Forage Quality of Tall fescue ( Festuca arundinacea Schreb.))

  • 이종경;김동암;이성철;최기춘;정종원;정재록
    • 한국초지조사료학회지
    • /
    • 제17권4호
    • /
    • pp.363-370
    • /
    • 1997
  • This experiment was carried out to investigate the effect of the mixed rates of endophyte-he and -infected seed on the endophyte fkquency, dry matter yield and forage quality of tall fescue(Festuca arundinacea Schreb.) at the Experimental Field of Grassland and Forage Crops Division, National Livestock Research Institute, Suweon, from 1993 to 1994. The results obtained are summarized as follows: Endophyte frequency was increased with mixed rates of higher endophyte-infected seed, however, it was not different between mixed rates of endophyte-free and -infected seed. DM yield of tall fescue was not different between mixed rates of endophyte-fke and -infected seed, and CP, NDF, ADF and IVDMD contents of forage quality were not influenced by mixed rates of endophyte-he and -infected seed. However, the weed contents were slightly increased with high level of endophyte-free seed. The results demonstrated that endophyk-fke tall fescue did not seem to be greatly weak under poor conditions, if the weed contents were slightly increase.

  • PDF

FE model of electrical resistivity survey for mixed ground prediction ahead of a TBM tunnel face

  • Kang, Minkyu;Kim, Soojin;Lee, JunHo;Choi, Hangseok
    • Geomechanics and Engineering
    • /
    • 제29권3호
    • /
    • pp.301-310
    • /
    • 2022
  • Accurate prediction of mixed ground conditions ahead of a tunnel face is of vital importance for safe excavation using tunnel boring machines (TBMs). Previous studies have primarily focused on electrical resistivity surveys from the ground surface for geotechnical investigation. In this study, an FE (finite element) numerical model was developed to simulate electrical resistivity surveys for the prediction of risky mixed ground conditions in front of a tunnel face. The proposed FE model is validated by comparing with the apparent electrical resistivity values obtained from the analytical solution corresponding to a vertical fault on the ground surface (i.e., a simplified model). A series of parametric studies was performed with the FE model to analyze the effect of geological and sensor geometric conditions on the electrical resistivity survey. The parametric study revealed that the interface slope between two different ground formations affects the electrical resistivity measurements during TBM excavation. In addition, a large difference in electrical resistivity between two different ground formations represented the dramatic effect of the mixed ground conditions on the electrical resistivity values. The parametric studies of the electrode array showed that the proper selection of the electrode spacing and the location of the electrode array on the tunnel face of TBM is very important. Thus, it is concluded that the developed FE numerical model can successfully predict the presence of a mixed ground zone, which enables optimal management of potential risks.

육계사에서의 미생물 활성 토탄의 악취저감 효과 (Odor reduction effect of microbially activated peat in broiler houses)

  • 김규래;이상준;김태연;크리스디안티;술히 아우파;민현숙;고경찬;조호성;오연수
    • 한국동물위생학회지
    • /
    • 제45권2호
    • /
    • pp.111-116
    • /
    • 2022
  • The purpose of this study is to evaluate the reduction effect of microbially activated peat on odor generated by livestock farms. The odor gas was measured by stirring the livestock manure sample with the existing litter and the microbially activated peat (Healtha Peat) was developed by this research team. In outdoor farm experiment, the measurements were performed by comparing broilers farm using rice husks and microbially activated peat as litter. The weight, mortality, shipment date, and odor levels (NH3) were measured before and after experiment. The result showed that NH3 levels were reduced by 100% in the Healtha Peat mixed group, Healtha Peat and rice husks mixed group. In the peat mixed group, Healtha Peat and saw dust mixed group showed reduce value at 99.6% and 99.1%, respectively. However the rice husks mixed group and saw dust mixed group showed a relatively weak NH3 reduction effect with values of 57.5% and 84.8%, respectively. After 3 months, the Healtha Peat mixed group and Healtha Peat and rice husks mixed group showed the highest NH3 reduction effect persistence. In the outdoor farm experiment, NH3 was present in farms using rice husks, but not in farms using Healtha Peat. In farms using Healtha Peat, the mortality and NH3 were reduced by 75% and >90%, respectively. The average body weight increased 18% and resulted to 10% decrease in the shipping date. These results implied that microbially activated peat has a clear effect on farm NH3 reduction and affects the productivity of farm animals.

혼합주의 종류에 따른 치아의 부식능 차이 평가 (Difference of dental erosive potential according to the type of mixed drink)

  • 김영석
    • 한국산학기술학회논문지
    • /
    • 제21권1호
    • /
    • pp.739-744
    • /
    • 2020
  • 이번 연구의 목적은 pH, 적정산도, 형광소실도(△F)를 이용하여 혼합주의 부식위험성과 부식능에 대해 평가하고자 하였다. 실험군은 다음과 같이 분류하였다: 소주, 깔라만시+소주, 요구르트+소주, 콜라+소주, 에너지드링크+소주. 혼합주는 소주와 음료수를 7:3의 비율로 혼합하였다. pH와 적정산도(TA5.5, TA7.0; 음료의 pH를 5.5와 7.0까지 끌어 올리는데 필요한 1M NaOH의 양)는 교반하면서 pH미터기를 이용하여 측정하였다. 혼합주의 부식능은 우치를 1, 2, 4, 6시간 동안 침적하였다가 각 시간마다 꺼내어 quantitative light-induced fluorescence (QLF-D)를 이용하여 촬영, 분석하였다. 연구 결과, 혼합주는 평균 pH는 3.17 ± 0.50, 소주의 pH는 8.6 ± 0.01를 나타내었다. TA5.5와 TA7.0는 각각 0.5~18, 0.5~23.5의 분포를 나타내었다. 요구르트 혼합주를 제외한 나머지 혼합주의 △F는 시간이 지남에 따라 증가하였으며, 요구르트 혼합주에서는 부식이 나타나지 않았다. 결론적으로, 깔라만시 혼합주가 가장 강한 산성과 부식능을 보였으며, 요구르트가 가장 약한 부식능을 나타냈다. 혼합주는 치아 법랑질에 강한 부식위험성을 가지고 있으므로, 이에 대한 주의와 교육이 필요할 것으로 사료된다.

A General Mixed Linear Model with Left-Censored Data

  • Ha, Il-Do
    • Communications for Statistical Applications and Methods
    • /
    • 제15권6호
    • /
    • pp.969-976
    • /
    • 2008
  • Mixed linear models have been widely used in various correlated data including multivariate survival data. In this paper we extend hierarchical-likelihood(h-likelihood) approach for mixed linear models with right censored data to that for left censored data. We also allow a general random-effect structure and propose the estimation procedure. The proposed method is illustrated using a numerical data set and is also compared with marginal likelihood method.

Small Area Estimation via Nonparametric Mixed Effects Model

  • Jeong, Seok-Oh;Shin, Key-Il
    • 응용통계연구
    • /
    • 제25권3호
    • /
    • pp.457-464
    • /
    • 2012
  • Small area estimation is a statistical inference method to overcome the large variance due to the small sample size allocated in a small area. Recently some nonparametric estimators have been applied to small area estimation. In this study, we suggest a nonparametric mixed effect small area estimator using kernel smoothing and compare the small area estimators using labor statistics.

中部地方 各地의 大氣汚染潛在力에 관한 硏究 (A Study on the Air Pollution Potential in the Central Part of Korea)

  • 李鍾範
    • 한국대기환경학회지
    • /
    • 제7권1호
    • /
    • pp.41-47
    • /
    • 1991
  • Air pollution potentials of the 7 cities in the central part of Korea were obtained with the mean wind speed within the mixed layer and the mixed layer height calculated by the Jump Model. Seasonal variation of the afternoon mixed layer height in Seoul area shows that low in winter and high in summer. Annual mean of the morning air pollution potential was lowest in Incheon and highest in Wonju. On the other hand annual mean of the afternoon air pollution potential was lowest in Incheon and highest in Chuncheon. Relatively low air pollution potential in Incheon can be explained as high mixed layer height and the effect of sea breeze.

  • PDF

Improved Algorithm for Case-Deletion Diagnostic in Mixed Linear Models

  • Lee, Jang-Teak
    • Communications for Statistical Applications and Methods
    • /
    • 제7권3호
    • /
    • pp.677-686
    • /
    • 2000
  • Outliers may occur with respect to any of the random components in mixed linear models. We develop a use of simple, inexpensive updating formulas to consider the effect of case-deletion for mixed linear models. The method described here requires inversions of an n x n matrix, where n is the number of nonempty cells. A numerical example illustrates the use of computational formulas.

  • PDF