• Communications for Statistical Applications and Methods
• /
• 제22권3호
• /
• pp.241-253
• /
• 2015

In this paper we predict the track of typhoons using a Bayesian principal component regression model based on wind field data. Data is obtained at each time point and we applied the Bayesian principal component regression model to conduct the track prediction based on the time point. Based on regression model, we applied to variable selection prior and two kinds of prior distribution; normal and Laplace distribution. We show prediction results based on Bayesian Model Averaging (BMA) estimator and Median Probability Model (MPM) estimator. We analysis 8 typhoons in 2006 using data obtained from previous 6 years (2000-2005). We compare our prediction results with a moving-nest typhoon model (MTM) proposed by the Korea Meteorological Administration. We posit that is possible to predict the track of a typhoon accurately using only a statistical model and without a dynamical model.

• Journal of the Korean Data and Information Science Society
• /
• 제13권2호
• /
• pp.271-284
• /
• 2002

In this paper we consider the robust inference for the parameter of linear regression model based on weighted least squares. First we consider the sequential test of multiple outliers. Next we suggest the way to assign a weight to each observation $(x_i,\;y_i)$ and recommend the robust inference for linear model. Finally, to check the performance of confidence interval for the slope using proposed method, we conducted a Monte Carlo simulation and presented some numerical results and examples.

• Journal of the Korean Data and Information Science Society
• /
• 제21권6호
• /
• pp.1319-1325
• /
• 2010

Quantile regression provides a more complete statistical analysis of the stochastic relationships among random variables. Sometimes quantile functions estimated at different orders can cross each other. We propose a new non-crossing quantile regression method applying support vector median regression to restricted regression quantile, restricted support vector quantile regression. The proposed method provides a satisfying solution to estimating non-crossing quantile functions when multiple quantiles for high dimensional data are needed. We also present the model selection method that employs cross validation techniques for choosing the parameters which aect the performance of the proposed method. One real example and a simulated example are provided to show the usefulness of the proposed method.

• Journal of the Korean Statistical Society
• /
• 제24권2호
• /
• pp.349-360
• /
• 1995

This paper deals with the robustness properties of the minimum disparity estimation in linear regression models. The estimators defined as statistical quantities whcih minimize the blended weight Hellinger distance between a weighted kernel density estimator of the residuals and a smoothed model density of the residuals. It is shown that if the weights of the density estimator are appropriately chosen, the estimates of the regression parameters are robust.

• Communications for Statistical Applications and Methods
• /
• 제21권4호
• /
• pp.285-296
• /
• 2014

The quantile residual life function has been effectively used to interpret results from the analysis of the proportional hazards model for censored survival data; however, the quantile residual life function is not always estimable with currently available semi-parametric regression methods in the presence of heavy censoring. A parametric regression approach may circumvent the difficulty of heavy censoring, but parametric assumptions on a baseline hazard function can cause a potential bias. This article proposes a Bayesian semi-parametric regression approach for inference on an unknown baseline hazard function while adjusting for available covariates. We consider a model-based approach but the proposed method does not suffer from strong parametric assumptions, enjoying a closed-form specification of the parametric regression approach without sacrificing the flexibility of the semi-parametric regression approach. The proposed method is applied to simulated data and heavily censored survival data to estimate various quantile residual lifetimes and adjust for important prognostic factors.

• Journal of the Korean Statistical Society
• /
• 제29권2호
• /
• pp.201-217
• /
• 2000

The method of generalized estimating equations (GEE) has become very popular for the analysis of longitudinal data. We extend this work to the use of M-estimators; the resultant regression estimates are robust to heavy tailed errors and to outliers. The proposed method does not require correct specification of the dependence structure between observation, and allows for heterogeneity of the error. However, an estimate of the dependence structure may be incorporated, and if it is correct this guarantees a higher efficiency for the regression estimators. A goodness-of-fit test for checking the adequacy of the assumed M-estimation regression model is also provided. Simulation studies are conducted to show the finite-sample performance of the new methods. The proposed methods are applied to a real-life data set.

• Journal of the Korean Data Analysis Society
• /
• 제20권6호
• /
• pp.2721-2731
• /
• 2018

This study purposes to derive the predictive model for the cold tolerance of Brassica napus, using the data collected in the Tree Breeding Lab of Gyeongsang National University during July and August of 2016. Three Brassica napus samples were treated at each of low temperatures from $4^{\circ}C$ to $-12^{\circ}C$ by decrement of $4^{\circ}C$, step by step, and electrolyte leakage levels were measured at each stage. Electrolyte leakages were observed tangibly from $-4^{\circ}C$. We tried to fit the six nonlinear regression models to the electrolyte leakage data of Brassica napus: 3-parameter logistic model, baseline logistic model, 4-parameter logistic model, (4-1)-parameter logistic model, 3-parameter Gompertz model, and (3-1)-parameter Gompertz model. The baseline levels of the electrolyte leakage estimated by these models were 4.81%, 4.07%, 4.19%, 4.07%, 4.55%, and 0%, respectively. The estimated median lethal temperature, LT50, were $-5.87^{\circ}C$, $-6.31^{\circ}C$, $-6.05^{\circ}C$, $-6.35^{\circ}C$, $-4.98^{\circ}C$, and $-5.15^{\circ}C$, respectively. We compared and discussed the measures of goodness of fit to select the appropriate nonlinear regression model.

• 한국도로학회논문집
• /
• 제18권4호
• /
• pp.83-92
• /
• 2016

PURPOSES : The purpose of this study was to develop safety performance functions (SPFs) that use zero-inflated negative binomial regression models for urban intersections in central business districts (CBDs), and to compare the statistical significance of developed models against that of regular negative binomial regression models. METHODS : To develop and analyze the SPFs of intersections in CBDs, data acquisition was conducted for dependent and independent variables in areas of study. We analyzed the SPFs using zero-inflated negative binomial regression model as well as regular negative binomial regression model. We then compared the results by analyzing the statistical significance of the models. RESULTS : SPFs were estimated for all accidents and injury accidents at intersections in CBDs in terms of variables such as AADT, Number of Lanes at Major Roads, Median Barriers, Right Turn with an Exclusive Turn Lane, Turning Guideline, and Front Signal. We also estimated the log-likelihood at convergence and the likelihood ratio of SPFs for comparing the zero-inflated model with the regular model. In he SPFs, estimated log-likelihood at convergence and the likelihood ratio of the zero-inflated model were at -836.736, 0.193 and -836.415, 0.195. Also estimated the log-likelihood at convergence and likelihood ratio of the regular model were at -843.547, 0.187 and -842.631, 0.189, respectively. These figures demonstrate that zero-inflated negative binomial regression models can better explain traffic accidents at intersections in CBDs. CONCLUSIONS : SPFs that use a zero-inflated negative binomial regression model demonstrate better statistical significance compared with those that use a regular negative binomial regression model.

• Communications for Statistical Applications and Methods
• /
• 제16권5호
• /
• pp.781-790
• /
• 2009

본 논문에서는 퍼지수를 포함한 모수적 회귀모형을 추정하기 위하여 분포무관추정량으로 알려진 순위 변환방법과 Theil 방법을 소개한다. 순위 변환방법은 퍼지수의 ${\alpha}$-수준집합의 중심과 폭에 대한 순위를 이용하고 Theil 방법은 ${\alpha}$-수준집합의 중심과 폭에 대한 추정한 값들의 중위수를 이용한다. 예제를 이용하여 분포무관추정량으로 추정된 퍼지회귀모형의 효율성을 최소자승법과 여러 가지 방법으로 추정된 퍼지회귀모형과 비교한다.

• 한국도로학회논문집
• /
• 제12권2호
• /
• pp.43-49
• /
• 2010

본 연구는 청주시의 가로구간 교통사고를 다루고 있다. 연구의 목적은 가로구간의 사고모형을 개발하는 데 있다. 이를 위해서 이 연구에서는 전체 322개 세부구간으로 분리된 간선도로의 사고 자료를 이용하여 ZAM 모형을 개발하는데 중점을 두고 있다. ZAM 모형의 일종인 ZIP(zero inflated Poisson model)과 ZINB(zero inflated negative binomial model)를 중심으로 분석한 연구의 주요결과는 다음과 같다. 첫째, 모형의 적합성을 결정하는 Vuong 통계 값과 과분산계수 ${\alpha}$의 t 통계 값을 바탕으로 개발된 다양한 모형을 평가한 결과, 포아송, 음이항, ZIP 및 ZINB 회귀모형 중 ZINB 모형이 최적인 것으로 나타난다. 둘째, ZINB 모형은 t, ${\rho}$ 및 ${\rho}^2$값 (0.63)의 관점에서 보면, 다른 모형에 비해서 통계적으로 매우 의미 있는 모형으로 평가된다. 마지막으로, 개발된 ZINB 모형의 사고 요인은 교통량, 진출입구 수 그리고 중앙분리대 길이로 분석된다. 교통량과 진출입구 수는 사고발생에 '+'요인, 그리고 중앙분리대 길이는 '-'요인으로 평가된다.