• The Korean Journal of Applied Statistics
• /
• v.23 no.4
• /
• pp.683-692
• /
• 2010

This short article is concerned with a categorical time series obtained after clipping a heteroscedastic GARCH process. Estimation methods are discussed for the model parameters appearing both in the original process and in the resulting binary time series from a clipping (cf. Zhen and Basawa, 2009). Assuming AR-GARCH model for heteroscedastic time series, three data sets from Korean stock market are analyzed and illustrated with applications to calculating certain probabilities associated with the AR-GARCH process.

• The Korean Journal of Applied Statistics
• /
• v.8 no.2
• /
• pp.147-161
• /
• 1995

The primary objective of this paper is to review parametric models and test statistics related to overdspersion of count data. Poisson or binomial assumption often fails to explain overdispersion. We reviewed real examples of overdispersion in count data that occurred in toxicological or teratological experiments. We also reviewed several models that were suggested for implementing experiments. We also reviewed several models that were suggested for implementing the extra-binomial variation or hyper-Poisson variability, and we noted how these models were generalized and further developed. The approaches that have been suggested for the overdispersion fall into two broad categories. The one is to develop a parametric model for it, and the other is to assume a particular relationship between the variance and the mean of the response variable and to derive a score test staistics for detecting the overdispersion. Recently, Dean(1992) derived a general score test statistics for detecting overdispersion from the exponential family.

• The Korean Journal of Applied Statistics
• /
• v.27 no.2
• /
• pp.277-290
• /
• 2014

We analyzed binary landslide data from the Boeun area with logistic regression. Since the number of landslide occurrences is only 9 out of 5000 observations, this can be regarded as a rare event data. The main issue of logistic regression with the rare event data is a serious bias problem in regression coefficient estimates. Two bias correction methods were proposed before and we quantitatively compared them via simulation. Firth (1993)'s approach outperformed and provided the most stable results for analyzing the rare-event binary data.

• The KIPS Transactions:PartD
• /
• v.12D no.4 s.100
• /
• pp.657-662
• /
• 2005

The Non-Sampling Error can happen any time by means of the intended or unintended error by the interviewer or respondent, but it is very difficult to find the error in survey database because it can hardly be computed mathematically and systematically. Until now, we have found it accidentally through the simple relation between the items or through the inspection from the random field. Therefore we introduced an heuristic methodology that can detect the interviewer's error by statistical decision-making or data mining techniques with a case study. It will be helpful so as to improve the statistical duality and provide efficient field management for the supervisor.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.13 no.6
• /
• pp.54-63
• /
• 2014

Traffic accident frequency and severity level in Korea are known to be very serious. Especially the number of pedestrian fatalities was much worse and 1.6 time higher than the OECD average. According to the National Police Agency, the flash signals are reported to have many safety benefits as well as travel time reduction, which is opposed to the foreign studies. With this background of expanding the flash signal, this research aims to investigate the overall impact of the flash signal operation on safety, investigating and comparing the accident occurrence on the flash signal and the full signal intersections. For doing this accident prediction models for both flash and full signal intersections were estimated using independent variables (geometric features and traffic volume) and 3-year (2011-2013) accident data collected in Seoul. Considering the rare and random nature of accident occurrence and overdispersion (variance > mean) of the data, the negative binomial regression model was applied. As a result, installing wider crosswalk and increasing the number of pedestrian push buttons seemed to increase the safety of the flash signal intersections. In addition, the result showed that the average accident occurrence at the flash signal intersections was higher than at the full signal-operated intersections, 9% higher with everything else the same.

• Journal of the Korean Regional Science Association
• /
• v.38 no.1
• /
• pp.3-20
• /
• 2022

The purpose of this study is to exploratory analyze the transition pattern of establishments and workers in new growth industries in the metropolitan area from 2010 to 2019 and to identify regional factors affecting the inflow and outflow of new growth industry start-ups. As for the analysis, the original data of the Census on Establishments were used, and spatial data at the sigungu level were constructed based on the inflow and outflow data of the number of new growth industry businesses and workers. For the analysis, the degree centrality of connection to outflow inflow by region was calculated, and an empirical analysis was conducted on regional-level factors affecting the inflow and outflow of new growth industries by applying a negative binomial regression model. According to the results, the new growth industry manufacturing sector was actively relocated in southern Gyeonggi Province, and the new growth industry service sector in Gangnam and Guro-Geumcheon-gu, and the impact of regional-level factors on the inflow and outflow of new growth industry start-ups varies depending on the industry. This study presented implications for regional industrial policies to improve the competitiveness of the local economy by attracting new industries by identifying spatial transition patterns for new growth industries and conducting empirical analysis to identify influencing factors.

• The Korean Journal of Applied Statistics
• /
• v.29 no.2
• /
• pp.309-319
• /
• 2016

For binary classification models, we consider a risk score that is a function of linear scores and estimate the coefficients of the linear scores. There are two estimation methods: one is to obtain MLEs using logistic models and the other is to estimate by maximizing AUC. AUC approach estimates are better than MLEs when using logistic models under a general situation which does not support logistic assumptions. This paper considers imbalanced data that contains a smaller number of observations in the default class than those in the non-default for credit assessment models; consequently, the AUC approach is applied to imbalanced data. Various logit link functions are used as a link function to generate imbalanced data. It is found that predicted coefficients obtained by the AUC approach are equivalent to (or better) than those from logistic models for low default probability - imbalanced data.

• The Journal of the Korea Contents Association
• /
• v.20 no.7
• /
• pp.118-126
• /
• 2020

The purpose of the study is to analyze the relationship between the warning picture on a cigarette pack and non-smoking attempt, which is expected to contribute to the negative perception of smoking as a research subject about smoking adolescents. An online survey data of the Youth Health Behavior in 2018 has been used, and 3,722 adolescents who are currently smokers were selected for the study. For the measurement of variables, demographic sociology, health-related, and smoking-related factors have been revised, and multivariate binomial logistic regression analysis has been performed. The perception rate of cigarette warning pictures among adolescents who smoke currently is 84.7%, and among them, the attempt rate to quit smoking is 72.8%. As a result of the multivariate binomial logistic regression analysis, there is a meaningful relationship between adolescent smokers' attempts to quit smoking and whether they perceived cigarette pack warning pictures, and school grade year, academic performance, stress perception, and ease of purchasing cigarettes have been also expressed as meaningful variables. To be based on the result, it is necessary to manufacture to design a cigarette pack warning picture that can be easily recognized by smoking adolescents in the future.

• Journal of Korea Technology Innovation Society
• /
• v.15 no.3
• /
• pp.586-603
• /
• 2012

In order to relieve overcrowding, the Korean government has regulated firm's locations in the capital region of Korea. However, the standard of regulation mainly depends on the place of province. Using KIS-Value data of firms that are located Pyeong-taek(the Capital area) or Cheon-an(Non Capital area), in close proximity to each other, we utilize OLS and negative binomial regression models for identifying the difference of firms' management and innovation performance in terms of firms' location difference(regulation difference). Our analysis shows that innovation performance of firms in Cheon-an does better than Pyeong-taek's, but management performance has no gaps between them. This result indicates that the regulation of firm's location has influence on firm's innovation performance. Thus, regulation policy regarding firms' location need to be minutely amended.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.8 no.6
• /
• pp.46-54
• /
• 2009

This study deals with the effects on the accident reduction according to the installation of RLC (red light cameras). The objective is to analyze the effects on the accident reduction using EB (Empirical Bayes) method. In pursuing the above, the study uses the 728 accident data occurred at the 28 intersections which RLC are installed. The main results are as follows. First, the effects of accident reduction were analyzed to be 20.74% by simple before-after study method. Second, the safety performance functions (SPF) were developed by the Poisson and negative binominal regression models, and since the over-dispersion parameter was close to zero, Poisson model was evaluated to be more appropriate than the negative binominal model. Also, the Poisson model was analyzed to be statistically significant because its ${\rho}^2$ value was 0.409. Finally, the results of analysis using an EB method showed that the accidents were reduced by range from 3.89 to 29.23%.