• Communications for Statistical Applications and Methods
• /
• v.16 no.3
• /
• pp.463-477
• /
• 2009

Parameter estimation methods such as maximum likelihood estimation method, probability weighted moments method, regression method have been popularly applied to various extreme value models in numerous literature. Among three methods above, the performance of regression method has not been rigorously investigated yet. In this paper the regression method is compared with the other methods via Monte Carlo simulation studies for estimation of parameters of the Generalized Extreme Value(GEV) distribution and the Generalized Pareto(GP) distribution. Our simulation results indicate that the regression method tends to outperform other methods under small samples by providing smaller biases and root mean square errors for estimation of location parameter of the GEV model. For the scale parameter estimation of the GP model under small samples, the regression method tends to report smaller biases than the other methods. The regression method tends to be superior to other methods for the shape parameter estimation of the GEV model and GP model when the shape parameter is -0.4 under small and moderately large samples.

• Corrosion Science and Technology
• /
• v.17 no.2
• /
• pp.37-44
• /
• 2018

Taiwan has a typical marine climate featuring perennial high-temperature and dampness. This climate, together with the emission of various industrial corrosive waste gases in recent years, contributes a lot to the corrosion of metal materials. In this study, samples of carbon steel exposed to various atmospheres in Taiwan were analyzed to investigate the impacts of atmospheric factors on carbon steel corrosion. Carbon steel samples were collected from 87 experimental stations between 2009 and 2012. Statistical analysis was employed to investigate the correlations between the carbon steel corrosion situations and the atmospheric factors such as concentrations of sulfur dioxide or chloride, exposure time, rainfall, etc. The results indicate that for samples from industrial areas, the sulfur dioxide concentration and exposure time during fall and winter are significantly correlated to the condition of the carbon steel corrosion. However, for samples from coastal zones, the significant correlated factors are chloride concentration and wetting time during winter. The results of this study are useful for the development of carbon steel corrosion prediction models.

• Journal of Korean Society for Atmospheric Environment
• /
• v.9 no.2
• /
• pp.147-153
• /
• 1993

From March 1990 to August 1991, every each 5mm bulk precipitation samples were collected at one residental area in Taejon City to investigate chemical characteristics of acid rain. Major ion concentrations of rain samples $(pH, SO_4^{2-}, NO_3^-, CL^-, NH_4^+, Na^+, K^+, Ca^{2+}, Mg^{2+})$ were analysed and compared with the concentration of air pollutants (T. S. P, $SO_2, NO_x$) that were measured by Ministry of Environment. The results of statistical analysis are as followings. Rain pH was relatively high on October and January and relatively low on August, November and February. Major anion is sulfate, and it's concentration is 2.36 times higher than nitrate's, and major cations are ammonium, sodium and calcium ion. Monthly variation of sulfate and calcium concentrations are higher than the others. Ion concentration and rain pH were correlated negatively with rainfall amount. Major ions in rain samples were $SO_4^{2-}, NO_3^-, NH_4^+, Ca^{2+}$ and regression equations are proposed by multiple regression of measured data. Also, regression equation between air pollutants(T. S. P, $SO_2$) and $SO_4^{2-}, Na^+, K^+, Ca^{2+}, Mg^{2+} ions in rain samples were made. From this wer can predict rain pH.

• Smart Media Journal
• /
• v.9 no.3
• /
• pp.80-89
• /
• 2020

Out - of - distribution (OOD) samples are frequently encountered when deploying a classification model in plenty of real-world machine learning-based applications. Those samples are normally sampling far away from the training distribution, but many classifiers still assign them high reliability to belong to one of the training categories. In this study, we address the problem of removing OOD examples by estimating marginal density estimation using variational autoencoder (VAE). We also investigate other proper methods, such as temperature scaling, Gaussian discrimination analysis, and label smoothing. We use Chonnam National University (CNU) weeds dataset as the in - distribution dataset and CIFAR-10, CalTeach as the OOD datasets. Quantitative results show that the proposed framework can reject the OOD test samples with a suitable threshold.

• The Korean Journal of Health Service Management
• /
• v.8 no.2
• /
• pp.139-148
• /
• 2014

The purpose of this study was to analyze the factors affecting the unmet dental needs of Koreans. In relation to the subjects and methodology for this study, the data from the 5th national health nutrition survey which had been conducted between 2010 and 2012 were analyzed by using the PASW statistics 18.0 (SPSS Inc., Chicago, IL, USA), a statistical program, through the frequency analysis of complex samples, cross analysis of complex samples, and logistics regression analysis of complex samples. The results of analysis showed that the rate of unmet dental needs stood at 35.3% of all subjects. Moreover, to determine the factors related to the unmet dental needs, the logistics regression analysis of complex samples was conducted. The results of analysis showed that the demographic and socio-economic factors and oral health condition factors had relationship with unmet dental needs. It is necessary to consider all the objective dental treatments that reflect specific dental diagnosis from a comprehensive standpoint.

• Geomechanics and Engineering
• /
• v.14 no.4
• /
• pp.399-406
• /
• 2018

The influences of mineral content and porosity on ultrasonic wave velocity were assessed for ten hornfelsic rocks collected from southern and western parts of the city of Hamedan, western Iran. Selected rock samples were subjected to mineralogical, physical, and index laboratory tests. The tested rocks contain quartz, feldspar, biotite, muscovite, garnet, sillimanite, kyanite, staurolite, graphite and other fine grained cryptocrystalline matrix materials. The values of dry unit weight of the rocks were high, but the values of porosity and water absorption were low. In the rocks, the values of dry unit weight are related to the presence of dense minerals such as garnet so not affected by porosity. The statistical relationships between mineral content, porosity and ultrasonic wave velocity indicated that the porosity is the most important factor influencing ultrasonic wave velocity of the studied rocks. The values of P-wave velocity of the rocks range from moderate to very high. Empirical equations, relevant to different parameters of the rocks, were proposed to determine the rocks' essential characteristics such as primary and secondary wave velocities. Quality indexes (IQ) of the studied samples were determined based on P-wave velocities of them and their composing minerals and the samples were classified as non-fissured to moderately fissured rocks. Also, all tested samples are classified as slightly fissured rocks according to the ratio of S-wave to P-wave velocities.

• The Korean Journal of Applied Statistics
• /
• v.22 no.5
• /
• pp.1047-1057
• /
• 2009

The nonparametric bivariate two-sample test of Bennett (1967) is extended to the multivariate k sample test. This test has been easily modified for a monotone trend among k samples. Often in applications it is important to consider a set of multivariate response variables simultaneously, rather than individually, and also important to consider testing k samples altogether. Different approaches of estimating the null covariance matrices of the test statistics resulted in the same limiting form. The multivariate k sample test is applied to the non-normal data of a randomized trial conducted for a period of four weeks in mental hospitals. The purpose of the trial is to compare the efficacy of three different interventions for a relief of the frequently occurring problems of constipation, caused as a side effect of antipsychotic drugs during hospitalization. The bowel movement status of patient for a week is summarized into a single severity score, and severity scores of four weeks comprise a four-dimensional multivariate variable. It is desirable with this trial data to consider a multivariate testing among k samples.

• Journal of the Korean Society of Tobacco Science
• /
• v.17 no.1
• /
• pp.5-14
• /
• 1995

Efficient algorithms for the automatic classification of flue-cured tovacco ripeness and grading have been developed The ripeness of the tobacco was classified into 4 levels vased on the color. The lab-built simple RGB color measuring system was utilized for detecting the light reflectance of the tobacco leaves. The measured data were used far training the artificial neural network The performance of the trained network was also tested far the untrained samples. The spectrophotometer was used to detect the light reflectance and absorption of the graded tobacco leaves in the frequency ranges of the visible light The measured data and the statistical analysis was performed to investigate the light characteristics of the graded samples. The measured data were obtained from samples of 5 different grades directly without considering the leaf positions. Those data were used far training the artificial neural network The performance of the trained network was also tested far the untrained samples. The neural network based sensor information processing showed successful results for grading of tobacco leaves.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.5
• /
• pp.2603-2609
• /
• 2014

Reliability analysis of mechanical systems requires statistical modeling of input random variables such as distribution function types and statistical parameters that affect the performance of the mechanical systems. Some random variables are correlated, but considered as independent variables or wrong assumptions on input random variables have been used. In this paper, joint distributions were modeled using copulas and Bayesian method from limited number of data. To verify the proposed method, statistical simulation tests were carried out for various number of samples and correlation coefficients. As a result, the Bayesian method selected the most probable copula types among candidate copulas even though the candidate copula shapes are similar for low correlations or the number of data is limited. The most probable copulas also yielded similar reliabilities with the true reliability obtained from a true copula, so that it can be concluded that the Bayesian method provides accurate statistical modeling for the reliability analysis.

• Journal of Korean Society for Atmospheric Environment
• /
• v.12 no.3
• /
• pp.323-332
• /
• 1996

Precipitation samples were collected by the wet-only sampling method at Seoul from September 1993 to June 1995. Sample were analysed for the anions $(NO_3^-, NO_2^-, SO_4^{2-}, Cl^-, and F^-)$ and cations $(Na^+, K^+, Ca^{2+}, Mg^{2+}, and NH_4^+)$ in addition to pH and electric conductivity. In order to establish the chemical analysis data of high quality, the assurance checks for analytical data of precipitation were performed by considering the ion balance and by comparing the measured conductivity with the calculated conductivity. As we applied the various assurance checking methods by the ion balance used until recently to a data set measured in this study, the f value expressed as $\Sigma C/\Sigma A$ was found to be not appropriate for the data screening. Also, the scattering plot between cations and anions in each sample was found to show the general tendency of ion balance but was proved to not quantitate the standard of data screening at a data set of samples of various concentration levels. The h value defined as (A-C)/C for C $\geq$ A and (A-C)/A for C < A was used to check the ion balance. However, the standard of data screening by h value must very in response to total ion concentration of samples. In this study, the quality assurance of chemical analysis data was checked by considering both the ion balance of evaluating by h value and the conductivity balance. Further the quality control was achieved by these quality assurance methods. As the result, 67 samples among total 77 were obtained as valid. As the central tendency value for a statistical summary in the analytical parametr of samples, the volume-weighted mean value was found to represent more the general chemistry of precipitation rather than the arithmetic mean. The volume-weighted mean pH was 5.0 and 25% of samples was less than this mean. The concentrations of sufate and nitrate in precipitation were 90.4 ueq/L and 32.4 ueq/L which made up 59% and 21% of all anions. The raion of $SO_4^{2-}/(NO_3^- + NO_2^-)$ in precipitation was 2.7, which indicates that the contributions of $H_2SO_4$ and $HNO_3$ to the acidity of precipitation are 70% and 30%, respectively.