• The Korean Journal of Applied Statistics
• /
• v.29 no.7
• /
• pp.1459-1473
• /
• 2016

Both longitudinal data and survival data are collected simultaneously in longitudinal data which are observed throughout the passage of time. In this case, the effect of the independent variable becomes biased (provided that sole use of longitudinal data analysis does not consider the relation between both data used) if the missing that occurred in the longitudinal data is non-ignorable because it is caused by a correlation with the survival data. A joint model of longitudinal data and survival data was studied as a solution for such problem in order to obtain an unbiased result by considering the survival model for the cause of missing. In this paper, a joint model of the longitudinal zero-inflated count data and survival data is studied by replacing the longitudinal part with zero-inflated count data. A hurdle model and proportional hazards model were used for each longitudinal zero inflated count data and survival data; in addition, both sub-models were linked based on the assumption that the random effect of sub-models follow the multivariate normal distribution. We used the EM algorithm for the maximum likelihood estimator of parameters and estimated standard errors of parameters were calculated using the profile likelihood method. In simulation, we observed a better performance of the joint model in bias and coverage probability compared to the separate model.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.22 no.2
• /
• pp.82-96
• /
• 2019

This study evaluated the status of watershed health in Geum River Basin by SWAT (Soil and Water Assessment Tool) hydrology and water quality. The watershed healthiness from watershed hydrology and stream water quality was calculated using multivariate normal distribution from 0(poor) to 1(good). Before evaluation of watershed healthiness, the SWAT calibration for 11 years(2005~2015) of streamflow(Q) at 5 locations with 0.50~0.77 average Nash-Sutcliffe model efficiency and suspended solid (SS), total nitrogen(T-N), and total phosphorus(T-P) at 3 locations with 0.67~0.94, 0.59~0.79, and 0.61~0.79 determination coefficient($R^2$) respectively. For 24 years (1985~2008) the spatiotemporal change of watershed healthiness was analyzed with calibarted SWAT and 5 land use data of 1985, 1990, 1995, 2000, and 2008. The 2008 SWAT results showed that the surface runoff increased by 40.6%, soil moisture and baseflow decreased by 6.8% and 3.0% respectively compared to 1985 reference year. The stream water quality of SS, T-N, and T-P increased by 29.2%, 9.3%, and 16.7% respectively by land development and agricultural activity. Based on the 1985 year land use condition. the 2008 watershed healthiness of hydrology and stream water quality decreased from 1 to 0.94 and 0.69 respectively. The results of this study be able to detect changes in watershed environment due to human activity compared to past natural conditions.

• Tuberculosis and Respiratory Diseases
• /
• v.53 no.3
• /
• pp.309-318
• /
• 2002

Background : Usual interstitial pneumonia (UIP) is a fatal progressive fibrotic disorder of the lung with unknown etiology and characterized by a poor response to conventional immunosuppressive therapy. The histologic hallmark of UIP is parchy distribution of subpleural fibrosis and fibroblastic foci(FBF) with interposed normal appearing lung. Because FBF is a collection of actively proliferating myofibroblasts, it can be a marker of activity and prognosis of UIP. However, there were contradictory reports about the correlation between the degree of FBF and survival. Therefore we performed this study to investigate the value of FBF as prognostic marker of UIP. Methods : This was a retrospective study on the 46 patients(M:F=33:13, mean age:$59{\pm}12$ years) with UIP diagnosed by the surgical lung biopsy at the Asan Medical Center, Seoul, Korea between 1990 and 2000 and had follow-up of more than a year. All the biopsy specimens were reevaluated and diagnosed as UIP according to the ATS/ERS classification. Semiquantitative grading of FBF(absent, 0; mild 1; moderate 2; marked 3) by the experienced pathologists who did not know the clinical findings were compared to the clinical data and the follow up course. Results : Thirteen patients(28.2%) died of UIP progression during the study period. The median survival time of all the subjects was 26 months after the biopsy. At the univariate analysis, FVC, $D_Lco$, smoking history and the grade of FBF were significantly related to the survial. The survival was longer in subjects with lesser degrees of FBF, higher DLco, higher FVC and history of smoking. However the multivariate analysis with Cox regression test showed the extent of FBF was the only independent prognostic marker of UIP. Conclusion : These data suggested that the extent of FBF on the surgical lung biopsy can be used as a prognostic marker of UIP.

• Korean Journal of Environment and Ecology
• /
• v.32 no.1
• /
• pp.105-117
• /
• 2018

This study conducted health assessment and multivariate vegetation analysis using the riparian vegetation index in 30 sites of the Geumgang mainstream and Mihocheon to obtain practical data on the river management of the Geumgang. The result showed that the number of plant communities was 54. The flora was 75 families, 185 genera, 243 species, 2 subspecies, 21 varieties, 2 varieties, and 268 taxa. The riparian vegetation index was 38.3 (3.3; G-D1 ~ 66.7; G-U2, G-U4, and G-M3), and the health of the rivers in this area was evaluated as normal (grade C). The health of rivers was the highest in the upper stream of Geumgang mainstream and lowest in the downstream of Geumgang mainstream. The relationship between riparian vegetation index and chlorophyll-a content was low. The riparian vegetation was divided into five groups of Digitaria ciliaris colony group, Salix gracilistyla colony group, Erigeron annuus colony group, the group dominated by Humulus japonicus, Salix koreensis, Miscanthus sacchariflorus, and Phragmites japonica colonies, and the group dominated by Conyza canadensis and Echinochloa crusgalli var. echinata colonies. They had the similar health conditions. The CCA analysis showed that the environmental factors affecting the distribution of vegetation were physical factors such as vegetation area, artificial structure area, waterway area, branch width, channel width, and bank height and the biological factors such as the number of species. As such, it is necessary to maintain the health condition through continuous monitoring where the health condition is high and to apply active measures such as ecological restoration where the health condition is low.