• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.14 no.1
• /
• pp.85-93
• /
• 2015

Previous studies have estimated crash prediction models with the fixed effect model which assumes the fixed value of coefficients without considering characteristics of each intersections. However the fixed effect model would estimate under estimation of the standard error resulted in over estimation of t-value. In order to overcome these shortcomings, the random effect model can be used with considering heterogeneity of AADT, geometric information and unobserved factors. In this study, data collections from 89 intersections in Daejeon and estimates of crash prediction models were conducted using the random and fixed effect negative binomial regression model for comparison and analysis of two models. As a result of model estimates, AADT, speed limits, number of lanes, exclusive right turn pockets and front traffic signal were found to be significant. For comparing statistical significance of two models, the random effect model could be better statistical significance with -1537.802 of log-likelihood at convergence comparing with -1691.327 for the fixed effect model. Also likelihood ration value was computed as 0.279 for the random effect model and 0.207 for the fixed effect model. This mean that the random effect model can be improved for statistical significance of models comparing with the fixed effect model.

• Korea Trade Review
• /
• v.41 no.3
• /
• pp.177-198
• /
• 2016

China has retained economic growth rate of average 9% for more than ten years recently after China introduced capitalistic market economy system in 1979 by Deng Xiaoping. China has attracted foreign direct investment for a long time because it has retained very high economic growth rate, low labor cost, and various policies for foreign investors. This paper tries to analyse the determinants of foreign direct investment inflow after reform-opening of China with empirical analysis methods utilizing each province·city's specific characteristics by using the panel data from 1985 to 2013. For the empirical analysis we use random effect model, fixed effect model, pooled OLS, and random coefficient model. The results by pooled OLS and random coefficient model are presented for the comparison with the main results in the process of research. The research shows the results by fixed effect model are better than those by random effect model after doing Hausman's test. The results shows that GRDP, capital stock, and telecommunication exert a positive relationship with foreign direct investment, while express way variable exerts a negative one. China's education level surprisingly does not attract foreign direct investment even though it is not at a critical level. Therefore, the Chinese government should try to increase national income level as it symbolizes market size; encourage domestic investment; and construct high quality telecommunication infrastructure.

• The Journal of the Korea Contents Association
• /
• v.21 no.2
• /
• pp.548-561
• /
• 2021

Manufacturing is a pillar industry for national economic growth. Analyzing the internal problems of manufacturing enterprises can solve the difficulties faced by manufacturing enterprises and improve the overall performance of manufacturing enterprises.This study selected 1,546 listed manufacturing companies in Shenzhen and Shanghai stock markets from 2009 to 2015, and empirically analyzed the relationship between fixed assets and corporate performance by using the fixed effect model and the two-way fixed effect model.The study finds 1) the scale of fixed assets has a negative effect on corporate performance. 2) the quality of fixed assets has a weak positive relationship with fixed assets. 3) the growth rate of fixed assets impacts corporate performance positively.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.5
• /
• pp.787-796
• /
• 2009

A random coefficient model is applied when times of the repeated measurements are not fixed in experiments with respect to the subjects. The procedures of the inference of a random coefficient model are same as those of a mixed model. Diet experimental data was used for applying the random coefficient model. Various random coefficient models are investigated for the experimental data, and are compared each other. Finally, optimal random coefficient model would be selected. It resulted from the analysis that for the fixed effect factor, the baseline, treatment, height, and time effect were very significant. The treatment effect of the diet foods and exercises were more effective in losing weight than the effect of the diet foods only. The fixed cubic time effect was very significant. The variance components corresponding to the subject effect, linear time effect, quadratic time effect, and cubic time effect of the random coefficients are all positive. When quartic time effect was added as random coefficients the model did not converge. Thus random coefficients up to the cubic terms was considered as the optimal model.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.10
• /
• pp.145-150
• /
• 2017

This study aims to analyze panel data using OECD Health data of 34 years to examine how significant the inequality of income is to the inequality of health. The data was from OECD's pooled Health data of 32 countries from 1980 to 2013. The process of determining analysis model was as follows; First, through the descriptive statistics, we examined averages and standard deviation of variables. Second, Lagrange multiplier test has done. Third, through the F-test, we compared Least squares method and Fixed effect model. Lastly, by Hausman test, we determined proper model and examined effective factor using the model. As a result, rather than Pooled OLS Model, Fixed Effect Model was shown as effective in order to consider the characteristics of individual in the panel. The results are as follows: First, as relative poverty rate(${\beta}=-19.264$, p<.01) grows, people's life expectancy decreases. Second, as the rate of smoking(${\beta}=-.125$, p<.05) and the rate of unemployment (${\beta}=-.081$, p<.01) grows, people's life expectancy decreases. Third, as health expenditure(${\beta}=.414$, p<.01) shares more amount of GDP and as the number of hospital beds(${\beta}=-.190$, p<.05) grows, people's life expectancy increases.

• The Journal of Korean Academy of Prosthodontics
• /
• v.46 no.3
• /
• pp.290-297
• /
• 2008

STATEMENT OF PROBLEM: Implant-supported fixed cantilever prostheses are influenced by various biomechanical factors. The information that shows the effect of implant number and position of cantilever on stress in the supporting bone is limited. PURPOSE: The purpose of this study was to investigate the effect of implant number variation and the effect of 2 different cantilever types on stress distribution in the supporting bone, using 3-dimensional finite element analysis. MATERIAL AND METHODS: A 3-D FE model of a mandibular section of bone with a missing second premolar, first molar, and second molar was developed. $4.1{\times}10$ mm screw-type dental implant was selected. 4.0 mm height solid abutments were fixed over all implant fixtures. Type III gold alloy was selected for implant-supported fixed prostheses. For mesial cantilever test, model 1-1 which has three $4.1{\times}10$ mm implants and fixed prosthesis with no pontic, model 1-2 which has two $4.1{\times}10$ mm implants and fixed prosthesis with a central pontic and model 1-3 which has two $4.1{\times}10$ mm implants and fixed prosthesis with mesial cantilever were simulated. And then, 155N oblique force was applied to the buccal cusp of second premolar. For distal cantilever test, model 2-1 which has three $4.1{\times}10$ mm implants and fixed prosthesis with no pontic, model 2-2 which has two $4.1{\times}10$ mm implants and fixed prosthesis with a central pontic and model 2-3 which has two $4.1{\times}10$ mm implants and fixed prosthesis with distal cantilever were simulated. And then, 206N oblique force was applied to the buccal cusp of second premolar. The implant and superstructure were simulated in finite element software(Pro/Engineer wildfire 2.0). The stress values were observed with the maximum von Mises stresses. RESULTS: Among the models without a cantilever, model 1-1 and 2-1 which had three implants, showed lower stress than model 1-2 and 2-2 which had two implants. Although model 2-1 was applied with 206N, it showed lower stress than model 1-2 which was applied with 155N. In models that implant positions of models were same, the amount of applied occlusal load largely influenced the maximum von Mises stress. Model 1-1, 1-2 and 1-3, which were loaded with 155N, showed less stress than corresponding model 2-1, 2-2 and 2- 3 which were loaded with 206N. For the same number of implants, the existence of a cantilever induced the obvious increase of maximum stress. Model 1-3 and 2-3 which had a cantilever, showed much higher stress than the others which had no cantilever. In all models, the von Mises stresses were concentrated at the cortical bone around the cervical region of the implants. Meanwhile, in model 1-1, 1-2 and 1-3, which were loaded on second premolar position, the first premolar participated in stress distribution. First premolars of model 2-1, 2-2 and 2-3 did not participate in stress distribution. CONCLUSION: 1. The more implants supported, the less stress was induced, regardless of applied occlusal loads. 2. The maximum von Mises stress in the bone of the implant-supported three unit fixed dental prosthesis with a mesial cantilever was 1.38 times that with a central pontic. The maximum von Mises stress in the bone of the implant-supported three-unit fixed dental prosthesis with a distal cantilever was 1.59 times that with a central pontic. 3. A distal cantilever induced larger stress in the bone than a mesial cantilever. 4. A adjacent tooth which contacts implant-supported fixed prosthesis participated in the stress distribution.

• Communications for Statistical Applications and Methods
• /
• v.23 no.2
• /
• pp.179-187
• /
• 2016

The fixed effects panel probit model faces "incidental parameters problem" because it has a property that the number of parameters to be estimated will increase with sample size. The maximum likelihood estimation fails to give a consistent estimator of slope parameter. Unlike the panel regression model, it is not feasible to find an orthogonal reparameterization of fixed effects to get a consistent estimator. In this note, a hierarchical Bayesian model is proposed. The model is essentially equivalent to the frequentist's random effects model, but the individual specific effects are estimable with the help of Gibbs sampling. The Bayesian estimator is shown to reduce reduced the small sample bias. The maximum likelihood estimator in the random effects model is also efficient, which contradicts Green (2004)'s conclusion.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.8
• /
• pp.3417-3426
• /
• 2012

This study aims to determine how ownership structure (share-holding ratio of insiders, foreigners) affects agent costs (the portion of asset efficiency or non-operating expenses) through empirical analysis. However, as existing studies on correlations between ownership structure and agent costs adopted Pooled OLS Model, this study focused on additionally formulating Fixed Effect Model and Random Effect Model aimed to reflect the time of data formation and corporate effects as study models based on verification results on the suitability of Pooled-OLS Model before comparative analysis for the purpose of improvement of credibility and statistical validity of the results of empirical analysis based on the premise that the Pooled OLS Model is not reliable enough to verify massive panel data. The data has been accumulated over 10 years from 1998 to 2007 after the IMF crisis hit the nation, from a subject 331 companies except for financial institutions. As a result of the empirical analysis, verification of the suitability of model has determined that the Random Effect Model is appropriate in terms of asset efficiency among agent costs items. On the other hand, the Fixed Effect Model is appropriate in terms of non-operating costs. As a result of the empirical analysis according to the appropriate model, no hypothesis adopted in the Pooled OLS Model has been accepted. This suggests that developing an appropriate model is more important than other factors for the purpose of generating statistically significant empirical results by showing that different empirical results are produced according to the type of empirical analysis.

• Journal of Environmental Science International
• /
• v.7 no.6
• /
• pp.773-782
• /
• 1998

The adsorption experiment of phenol(Ph) from aqueous solution on granular activated carbon was studied in order to design the fixed-bed adsorption column. The experimental data were analyzed by unsteady-state, one-dimensional heterogeneous model. Finite element method(FEM) was applied to analyze the sensitivity of parameter and to predict the fixed-bed adsorption column performance on operation variable changes. The prediction model showed similar effect to mass transfer and intraparticle diffusion coefficient changes suggesting that both parameter present mass transfer rate limits for GAC-phenol system. The Freundlich constants had a greater effect than kinetic parameters for the performance of fixed-bed adsorption column. FEM solution facilitated prediction of concentration history in solution and within adsorbent particle.

• Korea Trade Review
• /
• v.41 no.5
• /
• pp.373-389
• /
• 2016

This study mainly investigates the port SOC's impact on trade volume. In order to investigate the relationships between port SOC and trade volume, we did the empirical analysis using panel data regression and fixed effects model. The total period of 97 years and 1,082 ports' information were applied to panel data and regression model. According to the results, the coefficients of development of container berth, development of bulk berth, maintenance of port, the jetty facilities like breakwater have positive(+) impact on the dependent variable, the trade volume. Especially, the jetty facilities show a strongly positive impact on trade volume. On the other hand, the development of new port and navigation facilities like lighthouse have a negative(-) impact. In examining Hausman test and LR test, the fixed effect model is statistically more appropriate than the random effect model for this study.