• Journal of the Korean Data and Information Science Society
• /
• v.24 no.3
• /
• pp.659-667
• /
• 2013

If the regression model or the propensity model is correct, the unbiasedness of the estimator using doubly robust imputation can be guaranteed. Using a neural network instead of a logistic regression model for the propensity model, the estimators using doubly robust imputation are approximately unbiased even though both assumed models fail. We also propose a doubly robust estimator of ratio form using population information of an auxiliary variable. We prove some properties of proposed theory by restricted simulations.

• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.307-308
• /
• 2007

This paper proposes a discriminative training algorithm for estimating hidden Markov model (HMM) parameters. The proposed algorithm estimates the Parameters by minimizing the p-norm of log-likelihood difference (PLD) between the utterance probability given the correct transcription and the most competitive transcription.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.50 no.6
• /
• pp.37-47
• /
• 2008

Photovoltaic power generation is currently being recognized as one of the most popular sources for renewable resources over the country. Although it is also being adapted to rural area for may reasons, it is important to estimate the magnitudes of power outputs with reliable statistical methodologies, while applying historical daily solar energy data, for correct feasibility analysis. In this study, one of the well-known statistical methodologies is employed to define the appropriate probability distributions for monthly power outputs for the selected rural area, county of Seo-san, province of Chungnam. The results imply that the assumption of normal distributions for several months may lead to incorrect decision-making and therefore lead to the unreliable feasibility analysis. Generalized beta and triangular distributions were found to be superior to normal distribution, when describing monthly probability distributions for daily photovoltaic power. Based on the appropriate distributions resulted from this study, Monte Carlo simulation technique was also applied to provide additional flexible information for the relevant decision makers. This study found out new finding that the probability distributions should be considered to make planning of the photovoltaic power system in rural village unit, in order to give reasonable economic analysis to the decision makers.

• Journal of Gastric Cancer
• /
• v.15 no.4
• /
• pp.262-269
• /
• 2015

Purpose: Intraabdominal abscess is one of the most common reasons for re-hospitalization after gastrectomy. This study aimed to develop a model for estimating the probability of intraabdominal abscesses that can be used during the postoperative period. Materials and Methods: We retrospectively reviewed the clinicopathological data of 1,564 patients who underwent gastrectomy for gastric cancer between 2010 and 2012. Twenty-six related markers were analyzed, and multivariate logistic regression analysis was used to develop the probability estimation model for intraabdominal abscess. Internal validation using a bootstrap approach was employed to correct for bias, and the model was then validated using an independent dataset comprising of patients who underwent gastrectomy between January 2008 and March 2010. Discrimination and calibration abilities were checked in both datasets. Results: The incidence of intraabdominal abscess in the development set was 7.80% (122/1,564). The surgical approach, operating time, pathologic N classification, body temperature, white blood cell count, C-reactive protein level, glucose level, and change in the hemoglobin level were significant predictors of intraabdominal abscess in the multivariate analysis. The probability estimation model that was developed on the basis of these results showed good discrimination and calibration abilities (concordance index=0.828, Hosmer-Lemeshow chi-statistic P=0.274). Finally, we combined both datasets to produce a nomogram that estimates the probability of intraabdominal abscess. Conclusions: This nomogram can be useful for identifying patients at a high risk of intraabdominal abscess. Patients at a high risk may benefit from further evaluation or treatment before discharge.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.50 no.6
• /
• pp.27-35
• /
• 2008

This study aims to estimate the optimal probability distribution of daily electricity generation by wind power, in order to contribute in rural green-village planning. Wind power generation is now being recognized as one of the most popular sources for renewable resources over the country. Although it is also being adapted to rural area for may reasons, it is important to estimate the magnitudes of power outputs with reliable statistical methodologies while applying historical daily wind data, for correct feasibility analysis. In this study, one of the well-known statistical methodology is employed to define the appropriate statistical distributions for monthly power outputs for specific rural areas. The results imply that the assumption of normal distributions for many cases may lead to incorrect decision-making and therefore lead to the unreliable feasibility analysis. Subjective methodology for testing goodness of fit for normal distributions on all the cases in this study, provides possibilities to consider the other various types of statistical distributions for more precise feasibility analysis.

• Proceedings of the KIEE Conference
• /
• summer
• /
• pp.88-90
• /
• 2004

Data are a critical utility asset. Collecting correct data on site leads to accurate information. Data, when gathered with foresight & properly formatted, are useful to both existing database and easily transferable to newer, more comprehensive historical outage data. However, when investigating data items options, the task, can be an arduous one, often requiring the efforts of entire committees. This paper firstly discusses the KEPCO's past 10 years of historical outage data which include meterological data, and also by several elements of the National Weather Service, failure rate, outage duration, and probability classification, etc. Then, these collected data are automatically stored in an Outage Data Management System (ODMS), which allows for easy access and display. ODMS has a straight-forward and easy-to-use interface. It lets you to navigate through modules very easily and allows insertion, deletion or editing of data. In particular, this will further provide the KEPCO that not only helps with probabilistic security assessment but also provides a platform for future development of Probability Estimation Program (PEP).

• Journal of Distribution Science
• /
• v.11 no.12
• /
• pp.5-11
• /
• 2013

Purpose - The purpose of this study was to construct a turnover estimation model by investigating research by Park et al. (2006) on the market area of domestic distribution. The study investigated distribution by using a new tool for the turnover estimation technique. This study developed and discussed the turnover estimation technique of Park et al. (2006), applying it to a large-scale retailer in "D"city that was suitable for on-the-spot distribution. It constructed the new model in accordance with test procedures keeping to this retail business location, to apply its procedures to a specific situation and improve the turn over estimation process. Further, it investigated the analysis and procedures of existing turnover estimation cases to provide problems and alternatives for turnover estimation for a large-scale retailer in "D"city. Finally, it also discussed problems and scope for further research. Research design, data, and methodology - This study was conducted on the basis of "virtue" studies. In other words, it took into account the special quality of the structure of Korea's trade zones. The researcher sought to verify a sale estimate model for use in a distribution industry's location. The main purpose was to enable the sale estimate model (that is, the individual model's presentation) to be practically used in real situations in Korea by supplementing processes and variables. Results - The sale estimate model is constructed, first, by conducting a data survey of the general trading area. Second, staying within the city's census of company operating areas, the city's total consumption expenditure is derived by applying the large-scale store index. Third, the probability of shopping is investigated. Fourth, the scale of sales is estimated using the process of singularity. The correct details need to be verified for the model construction and the new model will need to be a distinct sale estimate model, with this being a special quality for business conditions. This will need to be a subsequent research task. Conclusions - The study investigated, tested, and supplemented the turnover estimation model of Park et al. (2006) in a market area in South Korea. Supplementation of some procedures and variables could provide a turnover estimation model in South Korea that would be an independent model. The turnover estimation model is applied, first, by undertaking an investigation of the market area. Second, a census of the intercity market area is carried out to estimate the total consumption of the specific city. Consumption is estimated by applying indexes of large-scale retailers. Third, an investigation is undertaken on the probability of shopping. Fourth, the scale of turnover is estimated. Further studies should investigate each department as well as direct and indirect variables. The turnover estimation model should be tested to construct new models depending on the type of region and business. In-depth and careful discussion by researchers is also needed. An upgraded turnover estimation model could be developed for Korea's on-the-spot distribution.

• Phonetics and Speech Sciences
• /
• v.2 no.1
• /
• pp.77-85
• /
• 2010

This paper proposes an efficient feature vector processing technique to guard the Speech Emotion Recognition (SER) system against a variety of noises. In the proposed approach, emotional feature vectors are extracted from speech processed by comb filtering. Then, these extracts are used in a robust model construction based on feature vector classification. We modify conventional comb filtering by using speech presence probability to minimize drawbacks due to incorrect pitch estimation under background noise conditions. The modified comb filtering can correctly enhance the harmonics, which is an important factor used in SER. Feature vector classification technique categorizes feature vectors into either discriminative vectors or non-discriminative vectors based on a log-likelihood criterion. This method can successfully select the discriminative vectors while preserving correct emotional characteristics. Thus, robust emotion models can be constructed by only using such discriminative vectors. On SER experiment using an emotional speech corpus contaminated by various noises, our approach exhibited superior performance to the baseline system.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.2
• /
• pp.53-64
• /
• 2011

In this paper, we propose a channel estimation of side information method based received motion vectors for distributed video coding. The proposed decoder estimates motion vectors of side information and transmits it to the encoder. As the proposed encoder generates side information which is the same to one in the decoder with received motion vectors, accuracy of side information of the decoder is assessed and it is transmitted to decoder. The proposed decoder can also estimate accurate crossover probability with received error information. As the proposed method conducts correct belief propagation, computational complexity of the channel decoder decreases and error correction capability is significantly improved with the smaller amount of parity bits. Experimental results show that the proposed algorithm is better in rate-distortion performance and it is faster than several conventional distributed video coding methods.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.22 no.3
• /
• pp.282-291
• /
• 2011

Adaptive modulation(AM) is an important technique to increase the system efficiency, in which transmitter selects the most suitable modulation mode adaptively according to channel state in the temporary and spatially varying communication environment. Fixed modulation on channels with varying signal-to-noise ratio(SNR) is that the bit-errorrate(BER) probability performance is changing with the channel quality. An adaptive modulation scheme can be designed to have a BER which is constant for all channel SNRs. The correct as well as fast and simple SNR estimation is required essentially for this adaptive modulation. In order to operate adaptive modulation system effectively, in this paper, we analyze the effect of SNR estimation performance to it through the average BER and data throughput. Applying SNR estimation based on auto-correlation of decision feedback signal and others to adaptive modulation system, we also confirm performance degradation or improvement of its which is decided by SNR estimation error at each transition point of modulation level. Since SNR estimation based on auto-correlation of decision feedback signal shows stable estimation performance for various quadrature amplitude modulation(QAM) comparatively, this can be reduced degradation than others at each transition point of modulation level.