• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.12
• /
• pp.2655-2663
• /
• 2011

Several algorithms have been developed to detect AF which rely either on the form of P waves or the based on the time frequency domain analysis of RR variability. However, locating the P wave fiducial point is very difficult because of the low amplitude of the P wave and the corruption by noise. Also, the time frequency domain analysis of RR variability has disadvantage to get the details of irregular RR interval rhythm. In this study, we describe an atrial fibrillation detection algorithm through non-linear analysis of irregular RR interval rhythm based on the variability, randomness and complexity. We employ a new statistical techniques root mean squares of successive differences(RMSSD), turning points ratio(TPR) and sample entropy(SpEn). The detection algorithm was tested using the optimal threshold on two databases, namely the MIT-BIH Atrial Fibrillation Database and the Arrhythmia Database. We have achieved a high sensitivity(Se:94.5%), specificity(Sp:96.2%) and Se(89.8%), Sp(89.62%) respectively.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.2
• /
• pp.349-356
• /
• 2009

In this study, we suggest the estimation method of the relative risk for the unemployment statistics of a small area such as si, gun, gu in Korea. The considered method are the usual pooled estimator, weighted estimator with the inverse of log-variance as weights, and the Jackknife estimator. And we compare with the efficiency of the three estimators by estimating the bias and mean square errors using real data from the 2002 Economically Active Population Survey of Gyeonggi-do. We compute the unemployed rate of male and female in small areas, and then estimate the common relative risk for the unemployed rate between male and female. Also, the stability and reliability of the three estimators for the common relative risk was evaluated using the RB(relative bias) and the RRMSE(relative root mean square error) of these estimators. Finally, the Jackknife estimator turned out to be much more efficient than the other estimators.

• The Korean Journal of Applied Statistics
• /
• v.16 no.1
• /
• pp.1-14
• /
• 2003

In this study, we suggest the method of small area estimation based on the Economically Active Population Survey (EAPS) data in producing unemployment statistics for the local self-government areas (LSGAs) within large areas. The small area estimators considered are design-based indirect estimators such as the synthetic and composite estimators. The jackknife mean square error was used as a measure of accuracy of such small area estimators. The total unemployed and jackknife mean square errors of the 10 LSGAs within the large area of ChoongBuk region are derived from the estimation procedure suggested in this study, using EAPS data of December 2000. The reliability of small area estimators was assessed using the relative bias values and relative root mean square errors of these estimators. We find that under the current Korean EAPS system, the composite estimator turns out to be much more stable than other estimators.

• Water for future
• /
• v.28 no.5
• /
• pp.219-233
• /
• 1995

A non-stationary multivariate model is selected in which the mean and variance of rainfall are not temporally or spatially constant. And the rainfall prediction system is constructed which uses the recursive estimation algorithm, Kalman filter, to estimate system states and parameters of rainfall model simulataneously. The on-line, real-time, multivariate short-term, rainfall prediction for multi-stations and lead-times is carried out through the estimation of non-stationary mean and variance by the storm counter method, the normalized residual covariance and rainfall speed. The results of rainfall prediction system model agree with those generated by non-stationary multivariate model. The longer the lead time is, the larger the root mean square error becomes and the further the model efficiency decreases form 1. Thus, the accuracy of the rainfall prediction decreases as the lead time gets longer. Also it shows that the mean obtained by storm counter method constitutes the most significant part of the rainfall structure.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.24 no.3
• /
• pp.271-279
• /
• 2006

Acoustic Emission (AE) technology is an effective nondestructive testing for continuous monitoring of defect formation and failures in structural materials. This paper presents a source location model using Acoustic Emission (AE) sensors in a Pre-Stressed Concrete (PSC) beam and the evaluation of the model was performed through lab experiments. 54 AE events were made on the surface of the 5m-PSC beam using a Schmidt Hammer and arrival times were measured with 7AE sensors. The source location f3r each event was estimated using least squares. The results were compared with actual positions and the RMSE (Root Mean Square Errors) was about 2cm.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.20 no.3
• /
• pp.262-276
• /
• 2018

Models to predict Leaf Wetness Duration (LWD) were evaluated using the observed meteorological and dew data at the 11 citrus orchards in Jeju, South Korea from 2016 to 2017. The sensitivity and the prediction accuracy were evaluated with four models (i.e., Number of Hours of Relative Humidity (NHRH), Classification And Regression Tree/Stepwise Linear Discriminant (CART/SLD), Penman-Monteith (PM), Deep-learning Neural Network (DNN)). The sensitivity of models was evaluated with rainfall and seasonal changes. When the data in rainy days were excluded from the whole data set, the LWD models had smaller average error (Root Mean Square Error (RMSE) about 1.5hours). The seasonal error of the DNN model had the similar magnitude (RMSE about 3 hours) among all seasons excluding winter. The other models had the greatest error in summer (RMSE about 9.6 hours) and the lowest error in winter (RMSE about 3.3 hours). These models were also evaluated by the statistical error analysis method and the regression analysis method of mean squared deviation. The DNN model had the best performance by statistical error whereas the CART/SLD model had the worst prediction accuracy. The Mean Square Deviation (MSD) is a method of analyzing the linearity of a model with three components: squared bias (SB), nonunity slope (NU), and lack of correlation (LC). Better model performance was determined by lower SB and LC and higher NU. The results of MSD analysis indicated that the DNN model would provide the best performance and followed by the PM, the NHRH and the CART/SLD in order. This result suggested that the machine learning model would be useful to improve the accuracy of agricultural information using meteorological data.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.3
• /
• pp.653-661
• /
• 2016

In this paper, we provide a suitable optimal exponent in the generalized Pythagorean theorem and propose to use the logistic model & the probit model to estimate the winning rate in Korean professional baseball league. Under a criterion of root-mean-square-error (RMSE), the efficiencies of the proposed models have been compared with those of the Pythagorean theorem. We use the team historic win-loss records of Korean professional baseball league from 1982 to the first half of 2015, and the proposed methods show slight outperformances over the generalized Pythagorean method under the criterion of RMSE.

• The Korean Journal of Applied Statistics
• /
• v.27 no.3
• /
• pp.419-430
• /
• 2014

An accurate prediction of electricity supply and demand is important for daily life, industrial activities, and national management. In this paper electricity sales is predicted by time series modelling. Real data analysis shows the transfer function model with cooling and heating days as an input time series and a pulse function as an intervention variable outperforms other time series models for the root mean square error and the mean absolute percentage error.

• Journal of the Korea Society for Simulation
• /
• v.27 no.3
• /
• pp.99-107
• /
• 2018

In this paper, we study target tracking in two dimensional space using a Extended Kalman filter(EKF), various Extended Kalman Filter and Cost-Reference Particle Filter(CRPF), which can effectively estimate the state values of nonlinear measurement equation. We introduce various Extended Kalman Filter which the Unscented Kalman Filter(UKF), the Central Difference Kalman Filter(CDKF), the Square Root Unscented Kalman Filter(SR-UKF), and the Central Difference Kalman Filter(SR-CDKF). In this study, we calculate Mean Square Error(MSE) of each filters using Monte-Carlo simulation with unknown noise statistics. Simulation results show that among the various of Extended Kalman filter, Square Root Central Difference Kalman Filter has the best results in terms of speed and performance. And, the Cost-Reference Particle Filter has an advantageous feature that it does not need to know the noise distribution differently from Extended Kalman Filter, and the simulation result shows that the excellent in term of processing speed and accuracy.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.3
• /
• pp.513-522
• /
• 2014

Using the assumption that the price of a stock follows a geometric Brownian motion with constant volatility, Black and Scholes (BS) derived a formula that gives the price of a European call option on the stock as a function of the stock price, the strike price, the time to maturity, the risk-free interest rate, the dividend rate paid by the stock, and the volatility of the stock's return. However, implied volatilities of BS method tend to depend on the stock prices and the time to maturity in practice. To address this shortcoming, we estimate the implied volatility function as a function of the strike priceand the time to maturity for data consisting of the daily prices for KOSPI200 call options from January 2007 to May 2009 using support vector regression (SVR), the multiple additive regression trees (MART) algorithm, and ordinary least squaress (OLS) regression. In conclusion, use of MART or SVR in the BS pricing model reduced both RMSE and MAE, compared to the OLS-based BS pricing model.