• 한국환경과학회지
• /
• 제14권11호
• /
• pp.1049-1056
• /
• 2005

[ $NO_2$ ] concentration characteristics of Busan metropolitan city was analysed by statistical method using hourly $NO_2$ concentration data$(1998\~2000)$ collected from air quality monitoring sites of the metropolitan city. 4 representative regions were selected among air quality monitoring sites of Ministry of environment. Concentration data of $NO_2$, 5 air pollutants, and data collected at AWS was used. Both Stepwise Multiple Regression model and ARIMA model for prediction of $NO_2$ concentrations were adopted, and then their results were compared with observed concentration. While ARIMA model was useful for the prediction of daily variation of the concentration, it was not satisfactory for the prediction of both rapid variation and seasonal variation of the concentration. Multiple Regression model was better estimated than ARIMA model for prediction of $NO_2$ concentration.

• 응용통계연구
• /
• 제21권3호
• /
• pp.441-453
• /
• 2008

프로스포츠 선수들의 연봉은 선수들의 개인 성적과 팀에 대한 기여도 등으로 결정된다는 가정하에 프로농구와 프로야구 선수들의 전년도 성적으로 다음해 연봉을 예측 분석하였다. 분석에 있어서 data visualization 기법을 통해 변수사이의 관계, 이상점 발견, 모형진단등을 하였다. 다중선형회귀 모형(Multiple Linear Regression)과 트리모형(Regression Tree)을 이용해서 자료를 분석하고 모델간 비교를 했으며, Cross-Validation을 이용해서 최적모델을 선택하였다. 특히, 자동으로 변수선택을 하는 stepwise regression방법을 그냥 사용하기보다는 먼저 설명변수들 사이의 관계나 설명변수와 반응변수 사이의 관계등을 조사하고 나서 이를 통해 선택된 변수들을 가지고 stepwise regression과 regression tree 방법론을 이용해서 적절한 변수 및 최종 모형을 선택하였다. 분석결과, 프로농구의 경우에는 경기당 득점, 어시스트, 자유투 성공수, 경력 등이 중요한 변수였고, 프로야구 투수의 경우에는 경력, 9이닝 당 삼진 수, 방어율, 피홈런 수 등이 중요한 변수였고, 프로야구 타자의 경우에는 경력, 안타 수, FA(자유계약)유무 여부 등이 중요한 변수였다.

• 한국작물학회지
• /
• 제50권4호
• /
• pp.268-275
• /
• 2005

Early predictions of crop yields call provide information to producers to take advantages of opportunities into market places, to assess national food security, and to provide early food shortage warning. The objectives of this study were to identify the most useful parameters for estimating yields and to compare two model selection methods for finding the 'best' model developed by multiple linear regression. This research was conducted in two 65ha corn/soybean rotation fields located in east central South Dakota. Data used to develop models were small temporal variability information (STVI: elevation, apparent electrical conductivity $(EC_a)$, slope), large temporal variability information (LTVI : inorganic N, Olsen P, soil moisture), and remote sensing information (green, red, and NIR bands and normalized difference vegetation index (NDVI), green normalized difference vegetation index (GDVI)). Second order Akaike's Information Criterion (AICc) and Stepwise multiple regression were used to develop the best-fitting equations in each system (information groups). The models with $\Delta_i\leq2$ were selected and 22 and 37 models were selected at Moody and Brookings, respectively. Based on the results, the most useful variables to estimate corn yield were different in each field. Elevation and $EC_a$ were consistently the most useful variables in both fields and most of the systems. Model selection was different in each field. Different number of variables were selected in different fields. These results might be contributed to different landscapes and management histories of the study fields. The most common variables selected by AICc and Stepwise were different. In validation, Stepwise was slightly better than AICc at Moody and at Brookings AICc was slightly better than Stepwise. Results suggest that the Alec approach can be used to identify the most useful information and select the 'best' yield models for production fields.

• 한국환경과학회지
• /
• 제16권11호
• /
• pp.1301-1312
• /
• 2007

The objective of this study is to propose a critical storm duration forecasting model on storm runoff in small river basin. The critical storm duration data of 582 sub-basin which introduced disaster impact assessment report on the National Emergency Management Agency during the period from 2004 to 2007 were collected, analyzed and studied. The stepwise multiple regression method are used to establish critical storm duration forecasting models(Linear and exponential type). The results of multiple regression analysis discriminated the linear type more than exponential type. The results of multiple linear regression analysis between the critical storm duration and 5 basin characteristics parameters such as basin area, main stream length, average slope of main stream, shape factor and CN showed more than 0.75 of correlation in terms of the multi correlation coefficient.

• Journal of Microbiology and Biotechnology
• /
• 제10권4호
• /
• pp.476-481
• /
• 2000

A linear correlation, by the stepwise multiple regression analysis, was found between the sensory test of Sigumjang aroma and the gas chromatographic data which were transformed with logarithm. GC data is the most objective method to evaluate Sigumjang aroma. A multiple correlation coefficient and a determination coefficient of more than 0.9 were obtained at the 9th and 13th steps, respectively. At step 31, the coefficient of determination level of 0.95 was attained. The accuracy of its estimation became higher as the number of the variables entered into the regression model increased. Over 90% of the Sigumjang aroma was explained by 13 compounds indentified on GC. The contributing proportion of the peak 26 was the highest followed by peaks 57 (9.27%), 29 (7.51%), 54 (6.01%), 8 (5.99%), 49 (4.97%), and 13 (4.11%).

• 한국운동역학회지
• /
• 제26권4호
• /
• pp.353-360
• /
• 2016

Objective: The aim of this study was to investigate the relationship between aiming patterns and scores in archery shooting. Method: Four (N = 4) elementary-level archers from middle school participated in this study. Aiming pattern was defined by averaged acceleration data measured from accelerometers attached on the body during the aiming phase in archery shooting. Stepwise multiple regression analysis was used to test whether a model incorporating aiming patterns from all nine accelerometers could predict the scores. In order to extract period of interest (POI) data from raw data, a Dynamic Time Warping (DTW)-based extraction method was presented. Results: Regression models for all four subjects are conducted with different significance levels and variables. The significance levels of the regression models are 0.12%, 1.61%, 0.55%, and 0.4% respectively; the $R^2$ of the regression models is 64.04%, 27.93%, 72.02%, and 45.62% respectively; and the maximum significance levels of parameters in the regression models are 1.26%, 4.58%, 5.1%, and 4.98% respectively. Conclusion: Our results indicated that the relationship between aiming patterns and scores was described by a regression model. Analysis of the significance levels, variables, and parameters of the regression model showed that our approach - regression analysis with DTW - is an effective way to raise scores in archery shooting.

• 대한산업공학회지
• /
• 제42권5호
• /
• pp.314-326
• /
• 2016

The purpose of variable selection techniques is to select a subset of relevant variables for a particular learning algorithm in order to improve the accuracy of prediction model and improve the efficiency of the model. We conduct an empirical analysis to evaluate and compare seven well-known variable selection techniques for multiple linear regression model, which is one of the most commonly used regression model in practice. The variable selection techniques we apply are forward selection, backward elimination, stepwise selection, genetic algorithm (GA), ridge regression, lasso (Least Absolute Shrinkage and Selection Operator) and elastic net. Based on the experiment with 49 regression data sets, it is found that GA resulted in the lowest error rates while lasso most significantly reduces the number of variables. In terms of computational efficiency, forward/backward elimination and lasso requires less time than the other techniques.

• 대한지역사회영양학회지
• /
• 제3권5호
• /
• pp.679-684
• /
• 1998

This study was conducted to investigate the determinants of obesity during adolescnece. A total of 726 adolescents living in rural areas in Korea had been observed for four years from 1992 to 1996 regarding their diet, sexual maturity, blood profile and physical growth. Stepwise multiple regression analysis was used to identify priorities fo the importance between the factors influencing obesity. The average nutrient intake over the three year period was higher than that of the Korean Recommended Dietary Allowances. The prevalence of obesity for the subjects based on BMI was 9.5%. Results of the stepwise multiple regression analysis showed that blood components and sexual maturity were more significant factors for determining the obesity than the dietary factors. The result may suggest that to understand obesity in children it is necessary to develop on analytical model for the children rather than using the existing analytical model developed mostly for adult patients of obesity. The model should include a wide range of variables such as diet, sexual maturity and changes in blood.

• 대한토목학회논문집
• /
• 제26권1B호
• /
• pp.39-50
• /
• 2006

홍수를 예측하기 위해서 국내 5대강 유역의 홍수통제소는 저류함수모형을 사용하고 있으며 현재까지 홍수예측에 대한 많은 연구가 이루어지고 있다. 이에 본 논문에서는 현재 홍수통제소에서 사용되고 있는 저류함수모형과 과거의 강우-수위 관계를 이용한 회귀분석(regression analysis), 그리고 인공신경망(artificial neural network)을 이용하여 홍수를 예측하고 이를 비교, 분석하고자 하였다. 저류함수모형의 경우는 홍수통제소의 대표매개변수와 보정된 최적(평균)매개변수를 적용하였다. 그리고 회귀분석과 인공신경망은 1995~2001년까지의 홍수사상 중 4개의 홍수사상을 선택하여 회귀계수를 구하고 역전파(backpropagation) 알고리즘을 사용하여 학습을 시켰다. 그 결과 저류함수모형의 경우 최적 매개변수를 이용하였을 때 기존의 홍수통제소에서 사용하고 있는 대표매개변수보다 예측이 개선되었으며, 회귀분석의 방법인 다중회귀분석, Robust 회귀분석, Stepwise 회귀분석을 이용한 홍수예측은 비교적 정확한 결과를 얻을 수 있었다. 역전파 알고리즘을 사용한 인공신경망의 경우도 회귀분석을 이용한 홍수예측보다는 다소 못하였지만 정확한 결과를 얻을 수 있었다.

• 산업공학
• /
• 제23권1호
• /
• pp.78-88
• /
• 2010

In this paper we develop a model for the performance evaluation of defense R&D test development projects based on analytic hierarchy process. First, evaluation indicators are collected through the related literature survey and a delphi inquiry method. Second, stepwise multiple linear regression is used for developing a hierarchical structure for analytic hierarchy process in the evaluation model, which can make the selected evaluation indicators of the hierarchical structure independent. Also we verify the effectiveness of proposed indicators of the performance evaluation by comparing with the existing evaluation indicators. The developed indicators for the performance evaluation is more reasonable and practical than the previous indicators on defense R&D test development projects.