한국의 연령-소득-자산의 결합분포의 시각화: 조건부 분위수 곡선 접근법 = Visualizing Joint Distribution of Age, Income, and Asset in Korea: A Conditional Quantile Curves Approach

다국어 초록 (Multilingual Abstract)

Because of its multi-dimensionality and complexity of correlation among covariates, analyzing a joint distribution has been a daunting task for many researchers. Many studies are satisfied with presenting correlation matrices among the covariates, or carrying out multiple regressions. In this paper, we present an intuitively clear way of visualizing a joint distribution of age, income, and asset in Korea. We non-parametrically estimate conditional quantile curves from the joint distribution and provide a way of interpreting such curves. A conditional quantile curves approach used here provides a visually clear pictures, and allows us to have intuitive understanding of what is going on in the joint distribution. The method presented in this paper can be combined with various multivariate regression analyses. Data were obtained from the Korean Household Finance and Welfare Survey (KHFWS), jointly conducted by Statistics Korea, the Financial Supervisory Service, and the Bank of Korea. We analyze for years 2017 and 2024.