• The Korean Journal of Applied Statistics
• /
• v.6 no.1
• /
• pp.67-78
• /
• 1993

A new diagnostic statistic for detecting outliers and influential observations in linear models is suggested and studied in this paper. The proposed statistic is a weighted sum of two measures; one is for detecting outliers and the other is for detecting influential observations. The merit of this statistic is that it is possible to distinguish outliers from influential observations. We have done some Monte-Carlo Simulation to find the probability distribution of this statistic.

• Science of Emotion and Sensibility
• /
• v.20 no.2
• /
• pp.149-160
• /
• 2017

In hypothesis testing, the interpretation of a statistic obtained from the data analysis relies on a probabilistic distribution of the statistic constructed according to several statistical theories. For instance, the statistical significance of a mean difference between experimental conditions is determined according to a probabilistic distribution of the mean differences (e.g., Student's t) constructed under several theoretical assumptions for population characteristics. The present study explored the logic and advantages of random-resampling approach for analyzing event-related potentials (ERPs) where a hypothesis is tested according to the distribution of empirical statistics that is constructed based on randomly resampled dataset of real measures rather than a theoretical distribution of the statistics. To motivate ERP researchers' understanding of the random-resampling approach, the present study further introduced a specific example of data analyses where a random-permutation procedure was applied according to the random-resampling principle, as well as discussing several cautions ahead of its practical application to ERP data analyses.

• Journal of Broadcast Engineering
• /
• v.26 no.2
• /
• pp.184-196
• /
• 2021

This paper proposes statistics adaptive linear regression-based object size prediction method for object detection. YOLOv2 and YOLOv3, which are typical deep learning-based object detection algorithms, designed the last layer of a network using statistics adaptive exponential regression model to predict the size of objects. However, an exponential regression model can propagate a high derivative of a loss function into all parameters in a network because of the property of an exponential function. We propose statistics adaptive linear regression layer to ease the gradient exploding problem of the exponential regression model. The proposed statistics adaptive linear regression model is used in the last layer of the network to predict the size of objects with statistics estimated from training dataset. We newly designed the network based on the YOLOv3tiny and it shows the higher performance compared to YOLOv3 tiny on the UFPR-ALPR dataset.

• Proceedings of the Korean Statistical Society Conference
• /
• 2000.11a
• /
• pp.219-225
• /
• 2000

데이터 마이닝 분야에서 널리 사용되고 있는 신경망은 최근 많은 통계인들의 관심을 끌고 있다. 그러나 범용 근사성(universal approximator)이라는 성질에도 불구하고 초기치에 따라 적합 결과가 크게 좌우되는 단점이 있다. 본 논문에서는 붓스트랩 표본을 통해 초기치를 발견하는 bumping 기법이 신경망 분야에서 사용되고 있는 무작위 탐색법 보다 더 정확하고 안정적인 초기치를 제공하여 주는가를 살펴 보았다.

• Proceedings of the Korean Statistical Society Conference
• /
• 2005.11a
• /
• pp.101-105
• /
• 2005

호남지역에 대한 대설특보 예보를 위한 통계모형 개발을 수행하였다. 일 신적설량에 따라 세법주(0: 비발생, 1: 대설주의보, 2: 대설경보)로 구분되는 순서형 자료 형태를 지니고 있다. 두가지 통계 모형(다등급 로지스틱 회귀모형, 신경회로망 모형)을 고려하였으며, 수치모델 출력자료를 이용한 역학-통계모형 기법의 하나인 MOS(model output statistics)를 적용하여 축적된 수치모델 예보자료와 관측치의 관계를 통계모형식으로 추정하여 예측모형을 개발하였다. 군집분석을 사용하여 훈련자료와 검증자료를 구분하였으며, 예보치 생성을 위하여 문턱치를 고려하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.456-456
• /
• 2012

한강유역에 위치한 247개의 강우계에서 관측된 강우 자료를 분석하여 Modified Bartlett-Lewis Retangular Pulse Model (MBLRPM)의 매개변수들을 산정하고, 이들의 지도를 작성한 후, 이들의 정확도 및 매개변수들의 시/공간적 변화 유형을 분석하였다. 이를 위한 첫번째 과정으로, 각 강우 게이지에 대해 MBLRPM의 매개변수에 사용되는 통계치 (각 달에 대한 1, 3, 12, 24시간 누적 수준에서의 평균, 분산, 자기 상관계수, 무강우 확률)들을 계산한다. 이 후, 격자화된 한강유역의 각 셀에 대하여 앞서 계산된 강우 통계치를 Ordinary Kriging 공간 보간법을 통하여 할당한다. 이 후, 각 셀에 할당된 강우 통계치를 사용하여 MBLRPM의 매개변수들을 산정하여 각 매개변수들의 지도를 각 달에 대하여 얻는다. 매개변수 지도를 사용하여 MBLRPM에 의해 생성된 강우 데이터들은 관측치의 통계치를 정확성있게 재현하였으며, 시/공간적 경향성을 분석한 결과, 강우세포의 지속기간과 관련된 매개 변수를 제외한 나머지 5개의 매개변수들은 확연한 공간적 경향성을 보인 한 편, 시간적 경향성은 잘 나타나지 않았다. 본 연구 결과는 매개변수 산정이 힘든 MBLRPM의 특성을 극복하게 해주어 가상 강우 생성을 용이하게 함으로써 강우에 영향을 받는 여러 종류의 연구 주제에 대해 불확실성 분석을 할 수 있게 한다는 점에서 의미를 가질 수 있다.

• The Korean Journal of Applied Statistics
• /
• v.14 no.2
• /
• pp.223-231
• /
• 2001

다변량분석 기법을 위해서는 자료가 정규성(normality)가정을 만족해야한다. 본 연구에서는 GUI환경에서 일변량 및 다변량자료의 정규성검정, 이상치제거 및 변수변환을 하는 시스템을 Visual Basic 언어로서 구축하여 사용자들이 보다 편리하게 사용할 수 있음을 소개 하고자 한다.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.4B
• /
• pp.319-328
• /
• 2009

This study quantified the data by applying the EDA techniques considering the data structure, and the results were then used for the frequency analysis. Although traditional methods based on the method of moments provide very sensitive statistics to the extreme values, the EDA techniques have an advantage of providing very stable statistics with their small variation. For the application of the EDA techniques to the frequency analysis, it is necessary to normalization transform and inverse-transform to conserve the skewness of the raw data. That is, it is necessary to transform the raw data to make the data follow the normal distribution, to estimate the statistics by applying the EDA techniques, and then finally to inverse-transform the statistics of transformed data. These statistics decided are then applied for the frequency analysis with a given probability density function. This study analyzed the annual maxima one hour rainfall data at Seoul and Pohang stations. As a result, it was found that more stable rainfall quantiles, which were also less sensitive to extreme values, could be estimated by applying the EDA techniques. This methodology may be effectively used for the frequency analysis of rainfall at stations with especially high annual variations of rainfall due to climate change, etc.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.10
• /
• pp.2562-2570
• /
• 1997

This paper deals with the design and performance of the adaptive equalizer using high order statistics in order to improve the transmission characteristics of multipath fading channel. The multipath propagational phenomenon occurred in digital radio transmission causes the distortion and ISI of receiving signal. These are main reasons to increase the bit error rate and degrade the performance of receivers. In this paper, the adaptive equalization algorithm using high order statistics of received signal is used instead of CMA algorithm, Bussgang and Godard which are known widely. The performance of this algorithm (residualisi, recovered constellation, calculation) is presented varing SNR. As the result of the computer simulation, equalizer algorithm using high order statistics is better than CMA in the range of low SNR, $10{\sim}20dB$. Therefore, considering the actual communication systems which use the range of $14{\sim}20$ SNR, the adaptive equalizer using high order statistics can be used in the real multipath fading environment.

• Journal of dental hygiene science
• /
• v.4 no.2
• /
• pp.85-90
• /
• 2004

The purpose of this study was to examine what made dental hygiene students choose their major subject and how much they were satisfied with that in an effort to pave the way for more successful guidance. The subjects in this study were 320 dental hygiene students from five colleges in Gyeonggi province, Incheon, Gangweon province, north Gyeongsang province and north Jeolla province. 32 freshmen and 32 sophomores each were selected from the colleges.