• Speech Sciences
• /
• v.9 no.2
• /
• pp.135-146
• /
• 2002

Generally a speaker verification system improves its system recognition ratio by regularizing log likelihood ratio, using a speaker model and its background speaker model that are required to be verified. The speaker-based cohort method is one of the methods that are widely used for selecting background speaker model. Recently, Gaussian-based cohort model has been suggested as a virtually synthesized cohort model, and unlike a speaker-based model, this is the method that chooses only the probability distributions close to basic speaker's probability distribution among the several neighboring speakers' probability distributions and thereby synthesizes a new virtual speaker model. It shows more excellent results than the existing speaker-based method. This study compared the existing speaker-based background speaker models and virtual speaker models and then constructed new virtual background speaker model groups which combined them in a certain ratio. For this, this study constructed a speaker verification system that uses GMM (Gaussin Mixture Model), and found that the suggested method of selecting virtual background speaker model shows more improved performance.

• Journal of Korean Society of Rural Planning
• /
• v.12 no.3 s.32
• /
• pp.39-42
• /
• 2006

The purpose of this study is to develop rural population model adapting cohort survival method with sift-share effects. Administrative district in this study is below Myun: about 2,000 population. Population data of rural area in 1990, 1995, and 2000 by age cohort were selected for applying developed model. Damping coefficient from population data was calculated as 7% and results applying this coefficient in rural population data below the error from 12% to 1.06%. In detail, most of cohorts fitted with developed model except from 15 to 29 age groups. Application result of small population area; DaesulMyun revealed that main factor of population change is not natural change but migration.

• Journal of the Korean association of regional geographers
• /
• v.17 no.1
• /
• pp.75-95
• /
• 2011

A birth cohort approach to the Household life-cycle model could be an alternative to cross-sectional data. In this study, each residential mobilities of birth cohorts' is traced by the cohort data from repeated cross-section in the case of Jinju city. Because of the differences in fertilities by era, the volume of each cohort as a consumer in housing has varied and the condition of housing stock also has changed as the time goes by. These changes in housing make not only age effect stressed in Rossi's model, but also cohort and period effect. Due to theses effects of time, every residential mobility trajectories of generations' is different especially in earlier life stages. As households get older, it is found that the age effect reduces and the probability of residential mobility is lower. As this result, the residential succession and filtering between the earlier and latter generations is weakened and the residential segregation could be happened by birth cohort.

• 한국노년학
• /
• v.29 no.4
• /
• pp.1413-1425
• /
• 2009

The objective of the work is to analyze the factors affecting on changes in filial piety responsibility expectation. For the analysis, this study focuses on the two factors-aging effect and cohort effect. This work analyzes the 4 wave Hallym Aging Panel Data with random intercept model. In the study cohort is divided by the criteria of birth year 1940. and the former cohort is called colony-war cohort and the latter cohort is called industrialization-democratization cohort. The results are in following. First, older cohort shows higher filial piety responsibility expectation score than younger cohort. Second, age shows no relationship with filial responsibility expectation score. Third, male and resident in rural area shows higher score. Forth income, year of schooling, and subjective health show negative relationship with responsibility score.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2005.10a
• /
• pp.456-460
• /
• 2005

The purpose of this study is to develop a estimation model of migration with only population data using the cohort-survival model which has been used for forecasting future population. The fluctuation of population can be bisected to the natural change which can be occurred by birth and death and the social change which means migration. The factors of the social change are usually very important for establishing rural policies. However, researches using migration data has limitations because the usage of them are restricted. For verifying a estimation model of migration, comparing estimated population in 2000 year and migration quantity between 1996 and 2000 of 25 gu with real values, using population data and death ratio from 1995 to 2000 of the 25 gu in Seoul. Result shows a reliable data that R-square of forecating population model is 0.9755 and migration is 0.9180. So these model are worth to estimate a population and migration quantity to restricted migration data.

• The Journal of the Korea Contents Association
• /
• v.17 no.9
• /
• pp.396-405
• /
• 2017

Main purpose of this study is to analyze determinants of household debt among middle and old individuals aged between 32 and 76 that include Korean baby-boomers(born between 1955 and 1963), using a HAPC (Hierarchical Age-Period-Cohort) model and Korean Welfare Panel Study 2006-2016. This study includes 86,056 individuals. Research findings indicate that aging and period effects have statistically significant relationships with household debt levels, however, cohort effects including a baby-boomer generation do not. While household debt increases by 3,530,000 Korean won as age increases by one year, the rate of increase in household debt reduces as individual ages. In addition, employment and health status at the individual level have significant effects on household debt levels. The unemployed are more likely than the employed to have high household debt levels while unhealthy people tend than healthy people to have high household debt levels.

• Journal of the Economic Geographical Society of Korea
• /
• v.23 no.2
• /
• pp.166-181
• /
• 2020

This paper attempts to compare and analyze the intensity, trend, and regional gap of income inequality, capitalizing upon the Age-Period-Cohort model which considers age, time and cohort effects, with the 1998-2018 Korea Labor Panel (KLIPS) survey data for respondents living in the Capital and Non-Capital Regions. The main analysis results are as follows. First, in the case of both cohort and age effects, those in their 50~60s, including the so-called baby boomers and '386 generation' living in the Capital Region, have relatively lower income inequality effect compared to that of other age groups and cohorts in the Non-Capital Region. Second, the micro-individual characteristics cannot be ignored to account for a regional gap in income inequality, but rather the effects of structural and institutional omitted variables and the social discrimination effects of individual characteristics variables are more significant in explaining it. Overall, intra-and inter-cohort income inequalities appear to overlap.

• The Korean Journal of Applied Statistics
• /
• v.28 no.4
• /
• pp.703-719
• /
• 2015

We propose an extension of the Lee and Jho (2015) mean reverting the two factor mortality model by incorporating a period-specific cohort effect. We found that the consideration of cohort effect improves the mortality fit of Korea male data above age 65. Parameters are estimated by the weighted least squares method and Metropolis algorithm. We also emphasize that the cohort effect is necessary to choose the base survival index to calculate longevity bond issue price. A key contribution of the article is the proposal and development of a method to calculate the longevity bond price to hedge the longevity risk exposed to Korea National Pension Services.

• Health Policy and Management
• /
• v.24 no.1
• /
• pp.35-46
• /
• 2014

Background: This paper describes the relationship and effect of health examination on personal medical cost by identifying the difference of the cost for medical care in physician visit between the population without and with health examination. Methods: After classifying into three cohorts in which, independent variables were designed according to the Andersen's behavioral model, the association of personal medical cost for medical care and prescription drugs which is dependent variable was analyzed by t-test and Mann-Whitney test for description and gamma regression model for inference. Results: In personal average medical cost, the population with health examination paid significantly more than without health examination, 11.6% more in cohort 2008, 26.6% more in cohort 2009, and 48.0% more in combined cohort. The odds ratio on medical expenditure of outpatients with health examination was 1.067, 1.126, 1.398 significantly in cohort 2008, 2009, and combined cohort respectively, comparing to the group without health examination. In independent variables, that is female, the elderly, never married, non-working, non-metropolitan, the higher family income, the smaller family size, people with disability, the people with chronic disease, and people with health examination have significantly being paid more tendency showing positive association with medical cost. Conclusion: This result showed that medical expenditure in physician visit has been increased after taking a health examination. Therefore reasonable limitation of getting preventive medical service is suggested to avoid medical shopping around and reduce being repeated health examination by unifying control to find out easily the clinical results from various medical facilities.

• Communications for Statistical Applications and Methods
• /
• v.8 no.3
• /
• pp.815-822
• /
• 2001

The aim of this study is to estimate the seroconversion date of the human immunodeficiency virus(HIV) infection for the HIV infected patients in Korea. Data are collected from two cohorts. The first cohort is a group of "seroprevalent" patients who were seropositive and AIDS-free at entry. The other is a group of "seroincident" patients who were initially seronegative but later converted to HIV antibody-positive. The seroconversion dates of the seroincident cohort are available while those of the seroprevalent cohort are not. Estimation of seroconversion date is important because it can be used to calculate the incubation period of AIDS which is defined as the elapsed time between the HIV infection and the development of AIDS. In this paper, a Weibull regression model Is fitted for the seroincident cohort using information about the elapsed time since seroconversion and the CD4$^{+}$ cell count.The seroconversion dates for the seroprevalent cohort are imputed on the basis of the marker of maturity of HIV infection percent of CD4$^{+}$cell count.unt.