• The Korean Journal of Applied Statistics
• /
• v.23 no.1
• /
• pp.13-28
• /
• 2010

Assuming a mixture distribution for credit evaluation studies, we discuss estimating threshold methods to minimize errors that default borrowers are predicted as non defaults or non defaults are regarded as defaults. A method by using statistical hypotheses tests, the most powerful test and generalized likelihood ratio test, for the probability density functions which are defined with the score random variable and the parameter space consisted of only two elements such as the default and non default states is proposed to estimate a threshold. And anther optimal thresholds to maximize classification accuracy measures of the accuracy and the true rate for ROC and CAP curves are estimated as equations related with these probability density functions. Three kinds of optimal thresholds in terms of the hypotheses testing, the accuracy and the true rate are obtained from normal random samples with various means and variances. The sums of the type I and type II errors corresponding to each optimal threshold are obtained and compared. Finally we discuss about their efficiency and derive conclusions.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.6
• /
• pp.737-745
• /
• 2008

Recent developments of process and measurement technology bring much interest to the online monitoring of process operations such as milling, grinding, broaching, etc. The objective of online monitoring systems is to detect process changes as early as possible. This is helpful in protecting facilities against unexpected failures and then preventing unnecessary loss. This paper investigates, when the process monitoring data are obtained as a profile, the monitoring performances of a statistical $T^2$-statistic and a feedforward neural network by using a wavelet transform. Numerical experiments using cutting force data presented by Axinte show that the proposed wavelet based $T^2$-test has an acceptable power in detecting profile changes. However, its operating characteristic is very sensitive to autocorrelation. On the contrary, compared with $T^2$-test, the neural network has more stable performance in the presence of autocorrelation. This indicates that an adaptive feature to analyze noises should be incorporated into the wavelet based $T^2$-test.

• The Korean Journal of Applied Statistics
• /
• v.22 no.5
• /
• pp.911-921
• /
• 2009

Receiver Operating Characteristic(ROC) and Cumulative Accuracy Profile(CAP) curves are two methods used to assess the discriminatory power of different credit-rating approaches. The points of optimal classification accuracy on an ROC curve and of maximal profit on a CAP curve can be found by using iso-performance tangent lines, which are based on the standard notion of accuracy. In this paper, we offer an alternative accuracy measure called the true rate. Using this rate, one can obtain alternative optimal threshold points on both ROC and CAP curves. For most real populations of borrowers, the number of the defaults is much less than that of the non-defaults, and in such cases the true rate may be more efficient than the accuracy rate in terms of cost functions. Moreover, it is shown that both alternative scores of optimal classification accuracy and maximal profit are the identical, and this single score coincides with the score corresponding to Kolmogorov-Smirnov statistic used to test the homogeneous distribution functions of the defaults and non-defaults.

• Journal of Animal Science and Technology
• /
• v.49 no.2
• /
• pp.195-202
• /
• 2007

The growth traits in swine are economically important, which are measured by gain in weight during test period or by age of days to certain weight. However, the difference in growth rate due to individual performance and also other factors occurs. The more reasonable estimation of the measurements of these traits provides the less error in genetic evaluation of pigs. The data from 1,576 heads being weighed periodically of Gyungnam-heugdon(Berkshire) were analyzed to estimate the growth curve which is used to decide average daily gain and days to 90kg. It may not be possible to directly compare accuracy between the conventional methods and the alternative methods. However, the alternative methods by growth curve would be superior to the conventional methods not only in theoretical background, but also in acceptability for diverse factors such as breed, sex and age. The theoretical superiority of the alternative methods comes from estimation at same age in daily gain and calculation of additional days from measuring date to days to 90kg by growth curve of individual. Also this can be easily adopted in a computer system according to breed and sex.

• Journal of Korean Society of Forest Science
• /
• v.57 no.1
• /
• pp.1-7
• /
• 1982

The simple logistic growth model on the logistic curve, being originally a kind of population growth curve has also been sometimes utilized to describe growth curves in herbaceous plants such as duckweed and sun-flowers. It has already been recognized that the agreement between the theoretical calculations and the empirical observations is quite satisfactory form a practical point of view. It remains, however, still doubtful whether the logistic curve could be applied to the growth or ordinary woody plants which is quite different in its character from that of herbaceous plants. In this study, the simple logistic model, being a basic tool of stand density management, is applied to yield data from pine stands in order to test the adequacy of the model An attempt of testing the significance of the fit is made by applying the Chi-square test.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.215-215
• /
• 2017

NRCS-CN 방법은 총 강우량으로부터 유출량을 계산하는 방법으로, 국내에서는 설계홍수량 산정 시 NRCS-CN 방법의 사용을 권장하고 있다. CN값은 토지이용 및 피복, 토양특성, 수문학적 조건(AMC)에 따른 함수로 결정할 수 있으나, 보통의 경우 미국의 National Engineering Handbook (NEH-4)에서 제시한 표를 활용한다. 그러나, 우리나라의 토지피복 및 토지이용 현황은 미국과 다르기 때문에 현실 조건을 반영한 조정이 필요함에도 불구하고, 충분한 관측 자료가 확보되지 않아 이러한 조정이 어려운 실정이다. NRCS-CN 방법에서는 결과 값이 총 강수량보다 CN에 크게 의존적이기 때문에 부정확한 CN 값의 산정은 큰 오차를 야기할 수 있다. 또한 소유역에서는 초기손실량이 설계홍수량 산정에 큰 영향을 미치지만 우리나라는 초기손실률을 20%의 고정된 값을 일괄적으로 적용하고 있으며, 이는 제주도와 같은 특수한 투수성 지층에서는 적합하지 않다는 지적을 받아왔다. 여러 선행연구에서 강수량과 CN 사이에는 특정 관계식이 존재하며, 고정된 CN 값이 아닌 강수량에 따라 변화하는 값을 적용하는 것이 기존의 NRCS-CN 방법보다 더 정확한 결과를 나타낸다는 것이 확인된 바 있다. 본 연구에서는 NRCS-CN 방법의 CN 값과 초기손실률을 유역에 적합하게 개선하기 위해서 기존의 NRCS-CN 모형에 점근 유출곡선지수방법(Asymptotic CN Regression Method)을 통해 산정된 CN값과 각기 다른 초기손실률(0.01, 0.05, 0.10, 0.20, 0.40)을 적용하여 개선된 총 8개의 모형을 한강 권역 소유역에 적용하였다. RMSE, MAE 및 R-square 등의 지표를 이용하여 모형 검정을 수행하였으며, 최적의 모형 및 미개변수를 선정하였다. 그 결과 기존의 NRCS-CN 방법보다 점근 유출곡선지수방법을 적용했을 때 더 작은 오차를 나타내는 것을 확인하였으며, 대부분의 유역에서 0.01 또는 0.05 등 기존보다 더 작은 초기손실률을 채택 시 실측값과 가장 적은 오차를 나타냈다.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.65 no.4
• /
• pp.496-507
• /
• 2020

Virus infections of apples result in lowered commercial qualities such as low sugar content, weakened tree vigor, and malformed fruits. An effective way to control viruses is to produce virus-free plants based on the development of an accurate and sensitive diagnostic method. In this study, real-time PCR assays based on SYBR Green I and TaqMan probes were developed for detecting ASGV, ASPV, and ApMV viruses. These methods can detect and quantify 10³ to 10¹¹ RNA copies/μL of each virus separately. Compared with methods with two different dyes, the SYBR Green I-based method was efficient for virus detection as well as for assay using the TaqMan probe. Field tests demonstrated that real-time PCR methods developed in this study were applicable to high-throughput diagnoses for virus research and plant quarantine.

• Analytical Science and Technology
• /
• v.18 no.5
• /
• pp.369-375
• /
• 2005

The differential pulse anode stripping voltammograms of some rare earth elements and their complexes with catechol have been investigated in various pH and electrolytes. In a 0.1 M LiCl and pH 5.3 solution, $Euv^{3+}$ and $Pr^{3+}$ showed a single oxidation peak at -0.2 V and the oxidation currents were linearly increased with the concentration of those ions. $Tm^{3+}$, $Tb^{3+}$, $Yb^{3+}$ and $Sm^{3+}$ showed two oxidation peaks at -0.5 V and -0.2 V and the oxidation currents at -0.5 V were increased with the concentration increase of those ions. The linear range of those calibration curves was in 1 ppm-10 ppm. In the case of voltammograms of catechol complexes of rare earth elements, $Tb^{3+}$-catechol and $Eu^{3+}$-catechol complex showed a single oxidation peak at -0.95 V and -0.65V, respectively and $Sm^{3+}$-catechol, $Pr^{3+}$-catechol, $Tm^{3+}$-catechol and $Yb^{3+}$-catechol complexes showed two oxidation peaks. The linear range of the calibration curves of those complex was 0.1 ppm~1.0 ppm.

• Journal of the Korean Chemical Society
• /
• v.46 no.6
• /
• pp.509-514
• /
• 2002

A method to determin Fe(II) and Fe(III) ion in aqueous solution by chemiluminescence method using a stopped flow system has been studied. The method is based on the increased chemiluminescence intensity with the addition of Fe(III) ion to a solution of lucigenin and hydrogen peroxide. The effects of KOH concentration, flow rate of reagents, $H_2O_2$ concentration and citric acid concentration used for the masking of Fe(II) ion on the chemilu-minescence intensity have been investigated. The calibration curve for total Fe was linear over the range from 1.0${\times}$$10^{-6}$ M to 1.0${\times}$$10^{-4}$M, coefficient of correlation was 0.996 and the detection limit was 1.0${\times}$$10^{-7}$M under the optimal exper-imental conditions of 4.0 M, 2.0 M, 3.5 mL/min for the concentration of $H_2O_2,$ KOH and flow rate of reagents, respec-tively. The calibration curve for Fe(Ⅲ) was linear over the range from 1.0${\times}$$10^{-6}$M to 1.0${\times}10^{-4}$ M, the coefficient of correlation was 0.997 and the detection limit was 5.0${\times}$$10^{-7}$M under the optimal experimental conditions.

• Journal of Embryo Transfer
• /
• v.22 no.3
• /
• pp.143-147
• /
• 2007

Records on 546 Korea native pigs for average daily gain (ADG), age at 70 kg (D70 kg) and backfat thickness (BF) made between 2001 and 2006 in herds on National Institutes of Animal Science in Korea were used to estimate genetic parameters. The data was analyzed by the DF-REML (Derivative-Free Restricted Maximum Likelihood) program of Boldman using a single-trait animal model. Heritabilities were 0.26, 0.09, and 0.29 for ADG, D70 kg and BF, respectively. The phenotypic correlations of ADG with D70 kg and BF were -0.71 and 0.30. The phenotypic correlation of D70 kg with BF was -0.15. The genetic correlations of ADG with D70 kg and BF were -0.11, 0.41, respectively. The genetic correlation of D70 kg with BF was -0.16. The data of weights and measurements on body length, body height and chest width after age at 11 months (days to 330) were shown scarcely less differences compare to data of age at 11 months.