• Communications for Statistical Applications and Methods
• /
• 제10권2호
• /
• pp.291-303
• /
• 2003

This work is proposed an alternative identification method of outlying cell which is one of important issues in categorical data analysis. One finds that there is a strong relationship between the location of an outlying cell and the corresponding parameter estimates of the well-fitted log-linear model. Among parameters of log-linear model, an outlying cell is affected by interaction terms rather than main effect terms. Hence one could identify an outlying cell by investigating of parameter estimates in an appropriate log-linear model.

• 응용통계연구
• /
• 제18권2호
• /
• pp.395-410
• /
• 2005

많은 수의 범주형 변수에 의한 대규모 분할표 분석을 위하여 차원축소(collapsibility) 성질을 이용한 분석 방법을 제안하였다. kullback-Leibler의 발산 측도(divergence measure)를 이용한 서로 완전한 연관을 갖는 변수그룹을 결정하는 방법을 제안하였으며, 제안된 방법에 의한 변수그룹은 주변 로그선형모형(marginal log-linear models)에 의하여 변수들간의 연관성을 식별할 수 있다. 제안된 방법의 적용 예로 데이터 마이닝에서 흔히 접할 수 있는 대규모 분할표 자료인 소비자들의 구매행위 분석을 위한 장바구니 자료의 분석 결과를 제시하였다.

• Communications for Statistical Applications and Methods
• /
• 제30권1호
• /
• pp.95-107
• /
• 2023

Purpose: The statistical analysis of point processes on linear networks is a recent area of research that studies processes of events happening randomly in space (or space-time) but with locations limited to reside on a linear network. For example, traffic accidents happen at random places that are limited to lying on a network of streets. This paper applies techniques developed for point processes on linear networks and the tools available in the R-package spatstat to estimate the intensity of traffic accidents in Leon County, Florida. Methods: The intensity of accidents on the linear network of streets is estimated using log-linear Poisson models which incorporate cubic basis spline (B-spline) terms which are functions of the x and y coordinates. The splines used equally-spaced knots. Ten different models are fit to the data using a variety of covariates. The models are compared with each other using an analysis of deviance for nested models. Results: We found all covariates contributed significantly to the model. AIC and BIC were used to select 9 as the number of knots. Additionally, covariates have different effects such as increasing the speed limit would decrease traffic accident intensity by 0.9794 but increasing the number of lanes would result in an increase in the intensity of traffic accidents by 1.086. Conclusion: Our analysis shows that if other conditions are held fixed, the number of accidents actually decreases on roads with higher speed limits. The software we currently use allows our models to contain only spatial covariates and does not permit the use of temporal or space-time covariates. We would like to extend our models to include such covariates which would allow us to include weather conditions or the presence of special events (football games or concerts) as covariates.

• 자원ㆍ환경경제연구
• /
• 제13권4호
• /
• pp.717-734
• /
• 2004

본 연구는 생물경제학(bioeconomics)분석에 있어서 중요한 위치를 차지하는 생물학적 성장모델에 대한 계량적 접근을 시도하였다. 세계적으로 어류에 관한 생물학적 성장모델은 여러 연구자들에 의해 추정된 바 있으나, 갑각류나 패류와 관련된 적정 성장함수의 추정은 어류에 비해 크게 연구되어 있지 않은 실정이다. 이에 몇몇 연구자들에 의해 사용된 공통된 성장함수들(Linear, Log reciprocal, Double log, Polynomial, Linear with Interactions)을 생산방식과 지역별 환경요인을 감안하여 한국 참가리비(Patinopecten yessoensis)의 성장을 추정하는 데 응용해 보았으며, 가장 적절한 모델은 계량적 분석을 통하여 도출하였다. 분석결과 Log reciprocal 형태의 성장함수가 참가리비류에 가장 적합한 모델로 선정되었으며, 본 결과는 경영자의 최적 생산시기를 결정하는 데 이용되는 생물경제학 모델에 유용하게 응용될 수 있을 것으로 사료된다.

• Agricultural and Biosystems Engineering
• /
• 제2권1호
• /
• pp.15-23
• /
• 2001

On the assumption that the voices of the cows are produced by the linear prediction filter, we characterized the cows’voices. The order of this filter was determined by examining the voice characteristics both in time and frequency domains. The proposed order of the linear prediction filter is 15 for modeling voice production of the cow. The characteristics of the amplitude envelope of the voice signal was investigated by analyzing the sequence of the short time variance both in time and frequency domains, and the new parameters were defined. One of the coefficients o the linear prediction filter generating the voice signal, the fundamental frequency, the slope of the straight line regressed from the log-log spectra of the short time variance and the coefficients of the linear prediction filter generating the sequence of the short time variance of the voice signal can differentiate the two cows.

• Educational Technology International
• /
• 제15권2호
• /
• pp.71-88
• /
• 2014

A variety of studies to predict students' performance have been conducted since educational data such as web-log files traced from Learning Management System (LMS) are increasingly used to analyze students' learning behaviors. However, it is still challenging to predict students' learning achievement in blended learning environment where online and offline learning are combined. In higher education, diverse cases of blended learning can be formed from simple use of LMS for administrative purposes to full usages of functions in LMS for online distance learning class. As a result, a generalized model to predict students' academic success does not fulfill diverse cases of blended learning. This study compares two blended learning classes with each prediction model. The first blended class which involves online discussion-based learning revealed a linear regression model, which explained 70% of the variance in total score through six variables including total log-in time, log-in frequencies, log-in regularities, visits on boards, visits on repositories, and the number of postings. However, the second case, a lecture-based class providing regular basis online lecture notes in Moodle show weaker results from the same linear regression model mainly due to non-linearity of variables. To investigate the non-linear relations between online activities and total score, RF (Random Forest) was utilized. The results indicate that there are different set of important variables for the two distinctive types of blended learning cases. Results suggest that the prediction models and data-mining technique should be based on the considerations of diverse pedagogical characteristics of blended learning classes.

• 한국인구학
• /
• 제26권1호
• /
• pp.143-170
• /
• 2003

지난 몇 십 년 간에 걸쳐 한국 사회에서 남성들은 물론 여성들의 교육수준은 급속하게 증가하였다. 본 연구는 이러한 배경을 염두에 두면서 여성들의 교육수준이 첫 직업획득에 어떠한 영향을 끼치며 또 그것은 어떠한 양상으로 변하였는지 살펴본다. 이를 위해 한국여성개발원에서 2001년에 수집한 $\ulcorner$제4차 여성취업실태조사 원자료$\lrcorner$를 이용해 노동시장에 처음으로 진입한 시점을 기준으로 1980년 이전, 1980년대, 그리고 1990년대 코호트를 구분하였다. 우선, 코호트별로 상이한 교육과 직업의 분포를 통제한 후 교육과 직업의 전반적인 연관성을 파악하기 위해 대수선형모형(log-linear model) 분석을 시도하였다. 그 결과 최근 코호트로 올수록 교육과 직업의 전반적인 연관성이 약화되는 것으로 나타난다. 이어 OLS 회귀분석을 통해 각각의 교육수준이 직업위세점수에 미치는 영향을 살펴보았으며, 분석결과는 교육수준간 직업위세점수 차이가 최근 코호트로 올수록 감소하였다는 것을 보여준다. 결론적으로 본 연구는 그간 한국 사회의 급속한 교육팽창의 맥락 하에서 여성의 교육과 직업의 연관성이 점점 약화되어왔다는 것을 보여준다.

• 한국농업기계학회:학술대회논문집
• /
• 한국농업기계학회 2000년도 THE THIRD INTERNATIONAL CONFERENCE ON AGRICULTURAL MACHINERY ENGINEERING. V.II
• /
• pp.196-203
• /
• 2000

On the assumption that the voices of the cows are produced by the linear prediction filter, we characterized the cows' voices. The order of this filter is determined by examining the voices characteristics both in time and frequency domains. The proposed order of the linear prediction filter is 15 for modeling voice production of the cow. The combination of the two parameters of the fundamental frequency, the slope of the straight line regressed from the log-log spectra of the amplitude-envelope and the only one coefficient involved in the linear prediction filter can differentiate the two cows.

• 한국화재소방학회논문지
• /
• 제28권2호
• /
• pp.69-75
• /
• 2014

본 연구에서는 방화발생에 영향을 미치는 요인을 도출하기 위하여 발생건수를 종속변수로 하고 경제 인구 사회적 요인을 독립변수로 하는 다중회귀분석을 실시하였다. 다중회귀분석은 선형함수, 준로그함수, 역준로그함수, 이중로그함수 4가지 함수형태에 대해 적용하였으며, 각 단계별로 변수의 선택과 제외를 고려하는 단계적선택 방식을 적용하였다. 다중공선성 문제와 자기상관 문제를 해결하기 위하여 분산확대지수(VIF)와 Durbin-Watson 계수 이용하였으며, 4가지 함수모형에 대하여 수정된 R 제곱(설명력) 값이 0.935 (93.5%)로 가장 값이 높고 통계적으로 유의한 선형함수모형을 최적의 모형으로 결정하고 모형에 대한 해석을 진행하였다. 선형함수모형 결과 방화발생에 영향을 미치는 요인은 범죄발생건수(0.829), 일반이혼율(0.151), 재정자주도(0.149), 소비자물가상승률(0.099) 순으로 도출되었다.

• 한국소음진동공학회논문집
• /
• 제21권6호
• /
• pp.562-569
• /
• 2011

In this study linear and nonlinear dynamic stability characteristics of a medium-size high-speed turbocharger, whose rotor is supported by two 3-lobe journal bearings, are analyzed to evaluate and identify the effects of its bearing design variables. The rotor has the rated speed of 40,500 rpm and maximum continuous speed of 45,000 rpm. At first, utilizing the linear stability analysis method, bearing designs of yielding stable or unstable LogDecs as small as possible are searched by manipulating with machined bearing clearances and preloads. As next, utilizing the nonlinear analysis method, limit cycles of the rotor responses at the rated and maximum continuous speeds are simulated to check their acceptances. Results have shown that for the turbocharger rotor-bearing system considered, the 3-lobe journal bearing design with a smaller machined clearance and a larger preload are preferred for the stable rotor responses. More importantly, since there exists a good correlation between the linear and nonlinear stability analysis results, it is concluded that firstly the linear stability analysis method may be applied to screen quickly the ranges of bearing designs for stable or least unstable solutions and then, lastly the nonlinear stability analysis method may be deployed to check an absolute motion stability in terms of the limit cycle.