• Journal of the Korean Data and Information Science Society
• /
• v.20 no.5
• /
• pp.895-903
• /
• 2009

The penalized partial likelihood based on restricted maximum likelihood method has been widely used for the inference of frailty models. However, the standard-error estimate for frailty parameter estimator can be downwardly biased. In this paper we show that such underestimation can be corrected by using hierarchical likelihood. In particular, the hierarchical likelihood gives a statistically efficient procedure for various random-effect models including frailty models. The proposed method is illustrated via a numerical example and simulation study. The simulation results demonstrate that the corrected standard-error estimate largely improves such bias.

• Journal of the Korean Data and Information Science Society
• /
• v.19 no.4
• /
• pp.1247-1253
• /
• 2008

In this paper, we propose a interval censored regression spline model with a variance function (non-constant variance that depends on a predictor). Simulation studies show our estimates from MCECM algorithm are consistent, but biased when the sample size is small because of boundary effects. Also, we examined how the distribution of $x_i$ affects the converging speed of these consistent estimates.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.3
• /
• pp.1021-1028
• /
• 2006

We shall propose ML, ordinary jackknife and biased reducing estimators of the parameter in the right truncated exponential distribution with an unidentified uniform outlier when the truncated point is unknown and their biases and MSE's are compared numerically each other in the small sample sizes.

• Journal of Labour Economics
• /
• v.26 no.3
• /
• pp.77-114
• /
• 2003

Analyses of the special data sets constructed from the National Longitudinal Survey of Youth and the Panel Study of Income Dynamics reveal that, compared with an annual wage measure, survey week wages are significantly counter-cyclically biased due to selecting workers with strong labor market attachment. We also find that survey week wages are more counter-cyclically biased in high-wage industries than in low-wage industries, that is, inter-industry gaps of survey week wages are counter-cyclically biased. Unlike existing longitudinal studies, the current study concludes that real wages are much more procyclical in high-wage industries than in low-wage industries, which is attributed to our adoption of annual wages that is less subject to the selectivity bias. Our finding is consistent with the empirical regularity that real wages are much more procyclical for men than for women, as men are overrepresented in industries with greater real wage procyclicalities. Overall, current results do not support the predictions of segmented labor market theories for the cyclicality of real wages.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.1 s.33
• /
• pp.59-66
• /
• 2005

The organization of the web is progressively more being used to improve search and analysis of information on the web as a large collection of heterogeneous documents. Most people begin at a Web search engine to find information. but the user's pertinent search results are often greatly diluted by irrelevant data or sometimes appear on target but still mislead the user in an unwanted direction. One of the intentional, sometimes vicious manipulations of Web databases is a intentionally biased web page like Google bombing that is based on the PageRank algorithm. one of many Web structuring techniques. In this thesis, we regard the World Wide Web as a directed labeled graph that Web pages represent nodes and link edges. In the Present work, we define the label of an edge as having a link context and a similarity measure between link context and target page. With this similarity, we can modify the transition matrix of the PageRank algorithm. By suggesting a motivating example, it is explained how our proposed algorithm can filter the Web intentionally biased web Pages effective about $60\%% rather than the conventional PageRank.

• Convergence Security Journal
• /
• v.23 no.1
• /
• pp.27-33
• /
• 2023

Digitalization due to COVID-19 has rapidly developed artificial intelligence-based voice recognition technology. However, this technology causes unfair social problems, such as race and gender discrimination if datasets are biased against some groups, and degrades the reliability and security of artificial intelligence services. In this work, we compare and analyze accuracy-based unfairness in biased data environments using VGGNet (Visual Geometry Group Network), ResNet (Residual Neural Network), and MobileNet, which are representative CNN (Convolutional Neural Network) models of artificial intelligence. Experimental results show that ResNet34 showed the highest accuracy for women and men at 91% and 89.9%in Top1-accuracy, while ResNet18 showed the slightest accuracy difference between genders at 1.8%. The difference in accuracy between genders by model causes differences in service quality and unfair results between men and women when using the service.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2010.04a
• /
• pp.35-37
• /
• 2010

To construct precision geoid model, the gravity data having equal distribution and quality is necessary. In previous study, however, the geoid model has low precision since the biased distributed gravity data and some unverified data has been used and the gap between land and ocean exists. Now, the airborne and land gravity data was collected by various survey and the ship-borne gravity data and altimeter data has been achieved. Therefore, the precision geoid model development would be possible. And the GPS/Leveling data obtained by NGII could be used for construction of hybrid geoid in Korea. In this study, the procedure of geoid construction based on airborne, land, ship-borne and altimeter data using Remove-Restore technique will be explained. And the verification of gravimetric geoid and hybrid geoid would be introduced.

• Journal of Labour Economics
• /
• v.33 no.2
• /
• pp.135-164
• /
• 2010

This paper compares the earnings effect of inter-industry technology differences between the self-employed and wage earners. It is assumed that primary skills utilized by the self-employed and paid workers differ in nature, and thus the earnings effect of technology differences and its skill-biasness also differ for each type of workers. For the empirical analysis. Heckman's two-stage method and quantile regressions are fitted to Korean panel data. The earnings effect of technology differences turns skill- biased for wage earners (job-specific skills), but prevails for all self-employed workers (entrepreneurial skills) regardless of their schooling level. This sectoral difference holds for each different quantile of earnings distribution.

• Asian-Australasian Journal of Animal Sciences
• /
• v.15 no.7
• /
• pp.925-931
• /
• 2002

Gibbs sampling algorithms were implemented to the multi-trait threshold animal models with any combinations of multiple binary, ordered categorical, and linear traits and investigate the amount of bias on these models with two kinds of parameterization and algorithms for generating underlying liabilities. Statistical models which included additive genetic and residual effects as random and contemporary group effects as fixed were considered on the models using simulated data. The fully conditional posterior means of heritabilities and genetic (residual) correlations were calculated from 1,000 samples retained every 10th samples after 15,000 samples discarded as "burn-in" period. Under the models considered, several combinations of three traits with binary, multiple ordered categories, and continuous were analyzed. Five replicates were carried out. Estimates for heritabilities and genetic (residual) correlations as the posterior means were unbiased when underlying liabilities for a categorical trait were generated given by underlying liabilities of the other traits and threshold estimates were rescaled. Otherwise, when parameterizing threshold of zero and residual variance of one for binary traits, heritability estimates were inflated 7-10% upward. Genetic correlation estimates were biased upward if positively correlated and downward if negatively correlated when underling liabilities were generated without accounting for correlated traits on prior information. Residual correlation estimates were, consequently, much biased downward if positively correlated and upward if negatively correlated in that case. The more categorical trait had categories, the better mixing rate was shown.

• Human Ecology Research
• /
• v.56 no.1
• /
• pp.1-14
• /
• 2018

Technology and Home Economics are associated with gender-related roles. In this respect, students' attitude toward these subjects may be influenced by gender equity awareness with attitudes that may perpetuate gender-biased images of subjects. This study examined the influence of gender equity awareness of middle school students on attitudes toward Technology and Home Economics. Data were collected through a survey to 442 students from eight purposively sampled middle schools in Seoul. Three gender equity awareness groups were identified through a cluster analysis: Equity in house work group (n=163), Traditional gender role group (n=102), and Equity in all areas group (n=152). The analyses of variances enabled an examination of the effects of gender and gender equity awareness. Differences were found among gender and gender equity awareness groups on attitudes toward Home Economics, but not toward Technology. Girls showed higher preference, higher perceived usefulness than boys, but with a lower importance for career preparation for Home Economics. Traditional gender role group scored the lowest on usefulness and importance for everyday life, yet highest on importance for career preparation. Equity in all areas group perceived lowest importance of Home Economics for career preparation. The results show that Home Economics is more strongly gender-typed than Technology, and that effort is needed to change the gender-biased image of the subject.