• The Korean Journal of Applied Statistics
• /
• v.29 no.1
• /
• pp.113-122
• /
• 2016

Longitudinal data often occur in prospective follow-up studies. Joint model for longitudinal data and failure time has been applied on several works. In this paper, we extend it to the case where longitudinal data involve informative observation time process as well as competing risks survival times. We use a likelihood approach and derive an EM algorithm to obtain maximum likelihood estimate of parameters. A suggested joint model allows us to make inferences for three components: longitudinal outcome, observation time process and competing risk failure time. In addition, we can test the association among these components. In this paper, liver cirrhosis patients' data is analyzed. The relationship between prothrombin times measured at irregular visiting times and drop outs is investigated with a joint model.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.6
• /
• pp.1103-1115
• /
• 2012

In this study, we studied the effect of onion on hyperlipidemia in terms of factors, such as body weight, liver weight, kidney weight, heart weight, blood glucose, total cholesterol, triglycerides, HDL-cholesterol, and LDL-cholesterol. The hyperlipidemia supplement was significantly effective on the liver weight, kidney weight, blood glucose, total cholesterol, triglycerides, and LDL-cholesterol with the fixed effect model. However, the liver weight, blood glucose, total cholesterol, and triglycerides were significantly decreased with the random effect model on the heterogeneous factors selected by Galbraith plot. The existence of publication bias was checked by using a funnel plot.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.4
• /
• pp.717-726
• /
• 2011

The present study was carried out to summarize the effect of green tea in the diabetic rats by meta-analysis related studies. The association measure to test effect of green tea was Hedges' standardized mean difference. In this particular fixed effect model, body weight was significantly increased. Also, blood glucose, triglycerides were significantly decreased. In this case of heterogeneous variable, random effect model was applied. In this model, body weight was significantly increased. Also, blood glucose was significantly decreased in green tea treated group. According to the Meta-regression analysis, duration of injection was not significant for variables.

• The Korean Journal of Applied Statistics
• /
• v.28 no.2
• /
• pp.353-360
• /
• 2015

Various statistical models have been proposed over the last decade for spatially correlated Gaussian outcomes. The spatial linear mixed model (SLMM), which incorporates a spatial effect as a random component to the linear model, is the one of the most widely used approaches in various application contexts. Employing link functions, SLMM can be naturally extended to spatial generalized linear mixed model for non-Gaussian outcomes (SGLMM). We review popular SGLMMs on non-Gaussian spatial outcomes and demonstrate their applications with available public data.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.1
• /
• pp.143-152
• /
• 2017

Restriction of completely randomization within a block can be handled by a split plot factorial design splitted by several plots. $3{\times}3$ split plot factorial design with two fixed main factors and one fixed block shows that powers of the rank transformed statistic for testing whole plot factorial effect and split plot factorial effect are superior to those of the parametric statistic when existing effect size is small or the remaining effect size is relatively smaller than the testing factorial effect size. Powers of the rank transformed statistic show relatively high level for exponential and double exponential distributions, whereas powers of the parametric and rank transformed statistic maintain similar level for normal and uniform distributions. Powers of the parametric and rank transformed statistic with two fixed main factors and one random block are respectively lower than those with all fixed factors. Powers of the parametric andrank transformed statistic for testing split plot factorial effect with two fixed main factors and one random block are slightly lower than those for testing whole plot factorial effect, but powers of the rank transformed statistic show comparative advantage over those of the parametric statistic.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.6
• /
• pp.1439-1448
• /
• 2013

The analysis of variance for randomized block design or two way classification data is well known. In this paper, particularly, we considered two stage nested design in which the levels of one factor is not identical for different levels of another factor. We investigate the structural properties of two stage nested design and the properties of error sum of squares for random effect model. For the application of two way nested design, we consider two-period crossover design which is used commonly for the equivalence test to bio-similar product. The confidence interval estimation of the difference of two population means in the crossover design is discussed based on statistical package SPSS.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.3
• /
• pp.413-421
• /
• 2011

The present study was carried out to summarize the effect of garlic in the hyperlipidemia rats by meta-analysis related studies. The association measure to test effect of garlic was the mean difference (MD). In this particular fixed-effect model of mean difference, body weight, liver weight, kidney weight and heart weight were significantly decreased (p < 0.05). Also, blood glucose, plasma total cholesterol, plasma triglycerides, LDL-cholesterol, liver cholesterol, liver triglycerides were significantly decreased. HDL-cholesterol was significantly increased. In this case of heterogeneous variable, random effect model was applied. In this model, liver weight, blood glucose, plasma total cholesterol, plasma triglycerides, LDL-cholesterol, liver cholesterol, liver triglycerides were significantly decreased. HDL-cholesterol was significantly increased. According to the meta-regression analysis, duration of injection was significantly for kidney weight, testis weight, plasma total cholesterol, plasma triglycerides, HDL-cholesterol, LDLcholesterol, liver cholesterol, liver triglycerides.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.4
• /
• pp.649-656
• /
• 2012

The present study was carried out to summarize the effect of caffeine in the lipid metabolic by meta-analysis. The association measure to test effect of caffeine was the Hedges's standardized mean difference (HG). In this particular fixed-effect model of Hedges's standardized mean difference, weight gain, heart weight, serum total lipid, serum triglycerides and liver triglycerides were significantly decreased (p < 0.05). Also, serum HDL cholesterol and serum LDL cholesterol were significantly increased. In this case of heterogeneous variable, random effect model was applied. In this model, weight gain, heart weight, serum total lipid, serum triglycerides, serum LDL cholesterol and liver triglycerides were significantly decreased in caffeine treated group. Also HDL-cholesterol was significantly increased in caffeine treated group.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.2
• /
• pp.401-410
• /
• 2009

Due to the characteristics of 3 ${\times}$ 3 Latin square design which is composed of two block effects and one main effect, powers of rank transformed statistic for testing the main effect are very superior to powers of parametric statistic without regard to the type of population distributions. By order of when all three effects are fixed, when on one block effect is random, when two block effects are random, the rank transform statistic for testing the main effect shows relatively high powers as compared with the parametric statistic. Further when the size of main effect is big with one equivalent size of block effect and the other small size of block effect, powers of rank transformed statistic for testing the main effect demonstrate excellent advantage to powers of parametric statistic.

• The Korean Journal of Applied Statistics
• /
• v.28 no.5
• /
• pp.933-949
• /
• 2015

Audience rating competition in the domestic drama market has increased recently due to the introduction of commercial broadcasting and diversification of channels. There is now a need for thorough studies and analysis on audience rating. Especially, a drama rating is an important measure to estimate advertisement costs for producers and advertisers. In this paper, we study the drama rating prediction models using various data mining techniques such as linear regression, LASSO regression, random forest, and gradient boosting. The analysis results show that initial drama ratings are affected by structural elements such as broadcasting station and broadcasting time. Average drama ratings are also influenced by earlier public opinion such as the number of internet searches about the drama.