• Title/Summary/Keyword: 로그 회귀분석

Search Result 92, Processing Time 0.028 seconds

Linear Regression-based 1D Invariant Image for Shadow Detection and Removal in Single Natural Image (단일 자연 영상에서 그림자 검출 및 제거를 위한 선형 회귀 기반의 1D 불변 영상)

  • Park, Ki-Hong
    • Journal of Digital Contents Society
    • /
    • v.19 no.9
    • /
    • pp.1787-1793
    • /
    • 2018
  • Shadow is a common phenomenon observed in natural scenes, but it has a negative influence on image analysis such as object recognition, feature detection and scene analysis. Therefore, the process of detecting and removing shadows included in digital images must be considered as a pre-processing process of image analysis. In this paper, the existing methods for acquiring 1D invariant images, one of the feature elements for detecting and removing shadows contained in a single natural image, are described, and a method for obtaining 1D invariant images based on linear regression has been proposed. The proposed method calculates the log of the band-ratio between each channel of the RGB color image, and obtains the grayscale image line by linear regression. The final 1D invariant images were obtained by projecting the log image of the band-ratio onto the estimated grayscale image line. Experimental results show that the proposed method has lower computational complexity than the existing projection method using entropy minimization, and shadow detection and removal based on 1D invariant images are performed effectively.

Theoretical and Empirical Issues in Conducting an Economic Analysis of Damage in Price-Fixing Litigation: Application to a Transportation Fuel Market (담합관련 손해배상 소송의 경제분석에서 고려해야 할 이론 및 실증적 쟁점: 수송용 연료시장에의 적용)

  • Moon, Choon-Geol
    • Environmental and Resource Economics Review
    • /
    • v.23 no.2
    • /
    • pp.187-224
    • /
    • 2014
  • We present key issues to consider in estimating damages from price-fixing cases and then apply the procedure addressing those issues to a transportation fuel market. Among the five methods of overcharge calculation, the regression analysis incorporating the yardstick method is the best. If the price equation relates the domestic price to the foreign price and the exchange rate as in the transportation fuel market, the functional form satisfying both logical consistency and modeling flexibility is the log-log functional form. If the data under analysis is of time series in nature, then the ARDL model should be the base model for each market and the regression analysis incorporating the yardstick method combines these ARDL equations to account for inter-market correlation and arrange constant terms and collusion-period dummies across component equations appropriately so as to identify the overcharge parameter. We propose a two-step test for the benchmarked market: (a) conduct market-by-market Spearman or Kendall test for randomness of the individual market price series first and (b) then conduct across-market Friedman test for homogeneity of the market price series. Statistical significance is the minimal requirement to establish the alleged proposition in the world of uncertainty. Between the sensitivity analysis and the model selection process for the best fitting model, the latter is far more important in the economic analysis of damage in price-fixing litigation. We applied our framework to a transportation fuel market and could not reject the null hypothesis of no overcharge.

Directional conditionally autoregressive models (방향성을 고려한 공간적 조건부 자기회귀 모형)

  • Kyung, Minjung
    • The Korean Journal of Applied Statistics
    • /
    • v.29 no.5
    • /
    • pp.835-847
    • /
    • 2016
  • To analyze lattice or areal data, a conditionally autoregressive (CAR) model has been widely used in the eld of spatial analysis. The spatial neighborhoods within CAR model are generally formed using only inter-distance or boundaries between regions. Kyung and Ghosh (2010) proposed a new class of models to accommodate spatial variations that may depend on directions. The proposed model, a directional conditionally autoregressive (DCAR) model, generalized the usual CAR model by accounting for spatial anisotropy. Properties of maximum likelihood estimators of a Gaussian DCAR are discussed. The method is illustrated using a data set of median property prices across Greater Glasgow, Scotland, in 2008.

Estimation of Tunnel Convergence Using Statistical Analysis (통계처리를 활용한 터널 내공변위의 분석에 관한 연구)

  • 김종우
    • Tunnel and Underground Space
    • /
    • v.13 no.2
    • /
    • pp.108-116
    • /
    • 2003
  • Measured convergence data of a tunnel were investigated by means of statistical and regression analysis, where the rock mass were mainly composed of andesite and granite. The rock mass around tunnel were classified by RMR method into five different ratings, and then convergence data which belong to individual ratings were statistically processed to find out the appropriate regression equations. Exponential equations were better coincided with measured data than logarithmic equations. As the number of rock mass rating was increased, the magnitude and standard deviation of convergence were increased. Final convergence data were also investigated to study the relevance with both maximum displacement rate and early measured convergence. Some brief results of their relevance are presented. For instance, the regression coefficient between final convergence and maximum displacement rate was turned out to be 0.87 for this studied tunnel.

Development and Validation of Multiple Regression Models for the Prediction of Effluent Concentration in a Sewage Treatment Process (하수처리장 방류수 수질예측을 위한 다중회귀분석 모델 개발 및 검증)

  • Min, Sang-Yun;Lee, Seung-Pil;Kim, Jin-Sik;Park, Jong-Un;Kim, Man-Soo
    • Journal of Korean Society of Environmental Engineers
    • /
    • v.34 no.5
    • /
    • pp.312-315
    • /
    • 2012
  • In this study, the model which can predict the quality of effluent has been implemented through multiple regression analysis to use operation data of a sewage treatment plant, to which a media process is applied. Multiple regression analysis were carried out by cases according to variable selection method, removal of outliers and log transformation of variables, with using data of one year of 2011. By reviewing the results of predictable models, the accuracy of prediction for $COD_{Mn}$ of treated water of secondary clarifiers was over 0.87 and for T-N was over 0.81. Using this model, it is expected to set the range of operating conditions that do not exceed the standards of effluent quality. In conclusion, the proper guidance on the effluent quality and energy costs within the operating range is expected to be provided to operators.

Improvement of Rating Curve Fitting Considering Variance Function with Pseudo-likelihood Estimation (의사우도추정법에 의한 분산함수를 고려한 수위-유량 관계 곡선 산정법 개선)

  • Lee, Woo-Seok;Kim, Sang-Ug;Chung, Eun-Sung;Lee, Kil-Seong
    • Proceedings of the Korea Water Resources Association Conference
    • /
    • 2008.05a
    • /
    • pp.1770-1773
    • /
    • 2008
  • 수위-유량 관계 곡선식에 포함되어져 있는 매개변수를 추정하기 위해 많이 사용되는 로그선형 회귀분석은 잔차의 비등분산성(heterocesdascity)을 고려하지 못하므로 본 연구에서는 의사우도추정법(Pseudo-likelihood Estimation, P-LE)에 의해 분산함수를 추정하고 이와 함께 회귀계수를 추정할 수 있는 방법을 제시하였다. 이 과정에서 제시된 회귀잔차를 최소화하기 위하여 SA(simulated annealing)이라는 전역 최적화 알고리즘을 적용하였다. 또한 수위-유량 관계 곡선식은 단면 등의 영향으로 인해 구간에 따라 각각 다르게 구축되어져야 하므로 이를 보다 객관적으로 판단하고 분리 위치를 정확히 추정하기 위하여 Heaviside 함수를 의사우도함수에 포함시켜 결과를 추정하도록 하였으며, 2개의 구간을 가지는 유량자료를 이용하여 제시된 방법의 합리성을 통계적으로 실험하였다. 이와 같이 통계적 실험을 통해 제시된 방법들이 기존 방법과 비교하여 가질 수 있는 장점을 파악하였으며, 제시된 방법들을 금강유역 5개 지점에서 대해 수행하여 효율성을 검증하였다.

  • PDF

Various Graphical Methods for Assessing a Logistic Regression Model (로지스틱회귀모형의 평가를 위한 그래픽적 방법)

  • Kim, Kyung Jin;Kahng, Myung Wook
    • The Korean Journal of Applied Statistics
    • /
    • v.28 no.6
    • /
    • pp.1191-1208
    • /
    • 2015
  • Most statistical methods are dependent on the summary statistic. However, with graphical approaches, it is easier to identify the characteristics of the data and detect information that cannot be obtained by the summary statistic. We present various graphical methods to assess the adequacy of models in logistic regression that include checking log-density ratio, structural dimension, marginal model plot, chi-residual plot, and CERES plot. Through simulation data, we investigate and compare the results of graphical approaches under diverse conditions.

An Analysis on Vehicle Accident Factors of Intersections using Random Effects Tobit Regression Model (Random Effects Tobit 회귀모형을 이용한 교차로 교통사고 요인 분석)

  • Lee, Sang Hyuk;Lee, Jung-Beom
    • The Journal of The Korea Institute of Intelligent Transport Systems
    • /
    • v.16 no.1
    • /
    • pp.26-37
    • /
    • 2017
  • The study is to develop safety performance functions(SPFs) for urban intersections using random effects Tobit regression model and to analyze correlations between crashes and factors. Also fixed effects Tobit regression model was estimated to compare and analyze model validation with random effects model. As a result, AADT, speed limits, number of lanes, land usage, exclusive right turn lanes and front traffic signal were found to be significant. For comparing statistical significance between random and fixed effects model, random effects Tobit regression model of total crash rate could be better statistical significance with $R^2_p$ : 0.418, log-likelihood at convergence: -3210.103, ${\rho}^2$: 0.056, MAD: 19.533, MAPE: 75.725, RMSE: 26.886 comparing with $R^2_p$ : 0.298, log-likelihood at convergence: -3276.138, ${\rho}^2$: 0.037, MAD: 20.725, MAPE: 82.473, RMSE: 27.267 for the fixed model. Also random effects Tobit regression model of injury crash rate has similar results of model statistical significant with random effects Tobit regression model.

Effect of Participant Activity of SNS Based Online Event on the Diffusion

  • Hong, Jae-Won;Kwak, Jun-Sik
    • Journal of the Korea Society of Computer and Information
    • /
    • v.26 no.2
    • /
    • pp.221-227
    • /
    • 2021
  • In this paper, we tried to explore factors influencing the diffusion of online events through SNS by analyzing the online footprint of consumers. To this end, log data of online events conducted by "C" beer brands were collected and analyzed. The analysis unit of log data was set for each one hour, and the analyzing method used descriptive and regression analysis. Results are as follows. First, factors influencing the diffusion of the view of SNS-based online events were like, friend used coupon, and friend size. In particular, the size of friends had the greatest impact on the diffusion, which again suggests the importance of social hubs in online events. Second, factors influencing the diffusion of the number of inflows were also like, friend used coupon, and size of friends. Third, it was found that the number of reply did not affect the diffusion of views and inflows. This study is meaningful that it suggested an alternative plan to increase the effect of online events by using real data.

Estimation of Occurrence Probability of Socioeconomic Damage Caused by Meteorological Drought Using Categorical Data Analysis (범주형 자료 분석을 활용한 사회경제적 가뭄 피해 발생확률 산정 : 충청북도의 적용사례를 중심으로)

  • Yu, Ji Soo;Yoo, Jiyoung;Kim, Min-ji;Kim, Tae-Woong
    • Proceedings of the Korea Water Resources Association Conference
    • /
    • 2021.06a
    • /
    • pp.348-348
    • /
    • 2021
  • 가뭄 연구의 궁극적 목표는 가뭄 발생의 메커니즘에 대한 이해를 높이고, 예측기술을 향상시켜 선제적 대응이 가능하도록 하는 것이다. 일반적으로 가뭄분석에 활용되는 가뭄지표는 연속형 변수로 간주하여 확률모형을 구축하지만, 가뭄상태와 가뭄피해 자료는 순서형 및 이산형 변수이므로 범주형 자료 분석 기법을 적용하는 것이 더 적절하다. 따라서 본 연구에서는 기상학적 가뭄과 피해발생 사이의 관계를 규명하기 위해 범주형 자료 분석 방법 중 로그선형(log-linear) 모형과 로지스틱(logistic) 회귀모형을 활용하였다. 가뭄피해 예측을 위한 가뭄 피해 정보를 수집하는 것은 매우 어려운 일이다. 가뭄의 영향으로 인해 발생할 수 있는 피해의 종류가 다양하며, 여러 분야의 이해관계자가 받아들이는 가뭄의 피해 양상이 다르기 때문이다. 본 연구에서는 국가가뭄정보포털(drought.go.kr)에서 충청북도의 가뭄피해현황 자료를 수집하였다. 30년(1991~2020년)동안 238개 읍면동 중 34개 행정구역에서 총 272건의 가뭄피해가 발생한 것으로 확인되었다. 표준강수지수(SPI)를 이용하여 분석된 지역별 연평균 가뭄발생횟수는 약 8.44회이며, 가뭄이 가장 많이 발생한 해는 2001년(평균 가뭄발생 18.7회)이었다. 강수의 부족으로 인해 발생하는 기상학적 가뭄이 사회경제적 피해를 야기하는 수문학적 가뭄으로 전이되기까지 몇 주에서 몇 달까지 시간이 소요된다. 이러한 관계를 파악하기 위해 가뭄피해 발생 여부를 예측변수, 가뭄피해 발생 이전의 가뭄상태를 설명변수로 설정하여 기상학적 가뭄 발생에 따른 가뭄피해 발생 확률을 산정하였다. 그 결과 가뭄피해 발생 당시의 가뭄상태보다 그 이전에 연속된 가뭄상태가 있을 경우 가뭄피해 발생 확률이 약 2.5배 상승하는 것으로 나타났다.

  • PDF