• Communications for Statistical Applications and Methods
• /
• 제28권5호
• /
• pp.521-536
• /
• 2021

As life expectancies increase continuously over the world, the accuracy of forecasting mortality is more and more important to maintain social systems in the aging era. Currently, the most popular model used is the Lee-Carter model but various studies have been conducted to improve this model with one of them being 6-parametric factor model (6-PFM) which is introduced in this paper. To this new model, long short-term memory (LSTM) and regularized LSTM are applied in addition to vector autoregression (VAR), which is a traditional time-series method. Forecasting accuracies of several models, including the LC model, 4-PFM, 5-PFM, and 3 6-PFM's, are compared by using the U.S. and Korea life-tables. The results show that 6-PFM forecasts better than the other models (LC model, 4-PFM, and 5-PFM). Among the three 6-PFMs studied, regularized LSTM performs better than the other two methods for most of the tests.

• Journal of the Korean Data and Information Science Society
• /
• 제24권6호
• /
• pp.1439-1448
• /
• 2013

확률화 블록계획법이나 교차된 이원분류표에 대한 분산분석과 오차제곱합의 성질은 널리 알려져있다. 본 논문에서는 인자에 따른 수준에 또 다른 인자들이 내포되어 있는 지분계획 특히 이단계 지분계획에 대하여 구조적 특징을 설명하고 오차제곱합의 성질에 대하여 살펴보았다. 또한 이단계 지분계획의 활용으로 크로스오버 계획을 소개하고 생물학이나 약학 등의 분야에서 많이 사용되는 동등성 검정의 신뢰구간 구축에 대하여 설명하였고 실제자료와 SPSS 통계 페키지를 이용하여 분석함으로서 응용성을 부각시켰다.

• 대한토목학회논문집
• /
• 제31권4B호
• /
• pp.315-322
• /
• 2011

본 연구에서는 임의시간 환산계수(Conversion Factor, CF)와 관련하여 거의 유일한 이론적 연구인 Weiss 모형을 자세히 검토하고, 그 문제점을 보완한 수정 Weiss 모형을 제안하였다. 아울러 강우의 특성을 반영하지 못하는 이들 모형의 한계를 극복하는 시도의 하나로서 강우의 시간분포 특성을 고려허여 CF를 추정하고 이를 실제 관측 자료에 근거한 CF와 비교하였다. 그 결과 수정 Weiss 모형을 근거로 산정한 CF 값은 강우 시간분포의 형태에 따라 다양한 값으로 나타났다. 특히, 중앙으로 집중된 강우 시간분포 형태에 대한 이론적인 CF 값이 실측 자료에 근거한 국내 외 CF 값들과 매우 유사한 것을 확인할 수 있었다.

• International Journal of Internet, Broadcasting and Communication
• /
• 제14권2호
• /
• pp.136-146
• /
• 2022

Some surface defects make the weak quality of steel materials. To limit these defects, we advocate a one-stage detector model RetinaNet among diverse detection algorithms in deep learning. There are several backbones in the RetinaNet model. We acknowledged two backbones, which are ResNet50 and VGG19. To validate our model, we compared and analyzed several traditional models, one-stage models like YOLO and SSD models and two-stage models like Faster-RCNN, EDDN, and Xception models, with simulations based on steel individual classes. We also performed the correlation of the time factor between one-stage and two-stage models. Comparative analysis shows that the proposed model achieves excellent results on the dataset of the Northeastern University surface defect detection dataset. We would like to work on different backbones to check the efficiency of the model for real world, increasing the datasets through augmentation and focus on improving our limitation.

• 정보학연구
• /
• 제9권3호
• /
• pp.77-89
• /
• 2006

The event study analyzes returns around event date at a time. Event study provides estimation periods and cumulative returns. Stock split announcements are generally associated with positive abnormal returns. It is important to investigate the responses of stocks to new information contained in the announcements of stock splits. So It is important to study the long term performance in the case of Stock Split. This Study forced to two approach method in evaluating the performance, the event time portfolio approach and calendar time portfolio approach. The event time portfolio approach exists the CAR model, BHAR model and WR model. And the calendar time portfolio approach has the 3 factor model, 4 factor model, CTAR model, and RATS model.

• Nuclear Engineering and Technology
• /
• 제37권6호
• /
• pp.595-604
• /
• 2005

A new instrument, an average BDFT (Birectional Flow Tube), was proposed to measure the flow rate in single and two phase flows. Its working principle is similar to that of the Pilot tube, wherein the dynamic pressure is measured. In an average BDFT, the pressure measured at the front of the flow tube is equal to the total pressure, while that measured at the rear tube is slightly less than the static pressure of the flow field due to the suction effect downstream. The proposed instrument was tested in air/water vertical and horizontal test sections with an inner diameter of 0.08m. The tests were performed primarily in single phase water and air flow conditions to obtain the amplification factor(k) of the flow tube in the vertical and horizontal test sections. Tests were also performed in air/water vertical two phase flow conditions in which the flow regimes were bubbly, slug, and churn turbulent flows. In order to calculate the phasic mass flow rates from the measured differential pressure, the Chexal drift-flux correlation and a momentum exchange factor between the two phases were introduced. The test results show that the proposed instrument with a combination of the measured void fraction, Chexal drift-flux correlation, and Bosio & Malnes' momentum exchange model could predict the phasic mass flow rates within a $15\%$ error. A new momentum exchange model was also proposed from the present data and its implementation provides a $5\%$ improvement to the measured mass flow rate when compared to that with the Bosio & Malnes' model.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2003년도 추계학술대회
• /
• pp.39-47
• /
• 2003

The Bayes factors with improper noninformative priors are defined only up to arbitrary constants. So, it is known that Bayes factors are not well defined due to this arbitrariness in Bayesian hypothesis testing and model selections. The intrinsic Bayes factor by Berger and Pericchi (1996) and the fractional Bayes factor by O'Hagan (1995) have been used to overcome this problems. This paper suggests intrinsic priors for testing the equality of two lognormal means, whose Bayes factors are asymptotically equivalent to the corresponding fractional Bayes factors. Using proposed intrinsic priors, we demonstrate our results with a simulated dataset.

• 한국지하수토양환경학회:학술대회논문집
• /
• 한국지하수토양환경학회 2001년도 추계학술발표회
• /
• pp.115-119
• /
• 2001

Contaminant retardation factor is derived from the colloidal and contaminant transport equations for a four-phase porous medium: an aqueous phase, two mobile colloidal phases, and a solid matrix. It is assumed that the contaminant sorption to solid matrix and colloidal particles and the colloidal deposition on solid matrix follow the linear isotherms. The behavior of the contaminant retardation factor in response to the change of model parameters is examined employing the experimental data of Magee et al. (1991) and Jenkins and Lion (1993). In the four-phase system, the contaminant retardation factor is determined by both the contaminant association with solid matrix and colloidal particles and the colloidal deposition on solid matrix. The contaminant mobility is enhanced when the affinity of contaminants to mobile colloids increases. In addition, as the affinity of colloids to solid matrix decreases, the contaminant mobility increases.

• Journal of the Korean Data and Information Science Society
• /
• 제14권3호
• /
• pp.661-671
• /
• 2003

The Bayes factors with improper noninformative priors are defined only up to arbitrary constants. So, it is known that Bayes factors are not well defined due to this arbitrariness in Bayesian hypothesis testing and model selections. The intrinsic Bayes factor by Berger and Pericchi (1996) and the fractional Bayes factor by O'Hagan (1995) have been used to overcome this problems. This paper suggests intrinsic priors for testing the equality of two lognormal means, whose Bayes factors are asymptotically equivalent to the corresponding fractional Bayes factors. Using proposed intrinsic priors, we demonstrate our results with real example and a simulated dataset.

• 대한치과보철학회지
• /
• 제18권1호
• /
• pp.37-47
• /
• 1980

The purpose of this study was to investigate the stresses in different proximal margins and to measure, quantitatively, the effect of different modifications in the design of preparations on the stresses using two-dimensional photoelasticity. Photoelastic stress analysis is based on the phenomenon, exhibited by most transparent solids, of becoming birefringent, or doubly refracting, when strained. Two birefringent materials were used in this study, PSM-1 and PSM-5 in .standard sheet ($10'{\times}10'{\times}\frac{1}{4}'$ thickness), PSM-1(polyester) was used for constructing the substructure, and PSM-5(epoxy resin) was used in making the restorations to be investigated. Two birefringent materials were used in the construction of composite photoelastic model. Seven variable models were constructed. The peripheral dimensions of all model were constant and the models represent an occlusomesial section of a lower posterior molar. Model 1 represents the knife edge margin (shoulderless), Model 2 represents the chamfer, Model 3 represents a rounded shoulder(no sharp angle between the axial wall and gingival floor), Model 4 represents a flat shoulder (axial wall is a $90^{\circ}$ angle to the gingival wall), Model 5 represents $+15^{\circ}$ angulation, Model 6 has a $-15^{\circ}$ angulation, and Model 7 is the same as Model 4 except that it has a $45^{\circ}$ bevel. Improved artificial stone was used to represent dental cement in luting the composite photoelastic model. Static loading procedures(100 pounds) were used at preplanned sites. The results were as follows; 1. The stresses in the proximal portion of all tested models were compressive in nature when the proximal shoulders were loaded vertically on the same proximal marginal ridge. 2. The round and chamfered preparations were the optimum designs in proximoocclusal restorations. They showed the lowest stress concentration factor, i.e. 2.16 and 2.23, respectively. The knife edged shoulder had the highest value, K=5.39. Round type shoulder geometry experiments reduced the stress concentration factor (S.C.F.) 3. The gingival portion of proximal shoulder geometry was a critical location for stress concentration.