The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Factors affecting regional population of Korea using Bayesian quantile regression

베이지안 분위회귀모형을 이용한 지역인구에 영향을 미치는 요인분석

  • Kim, Minyoung (Department of Statisctics, Ewha Womans University) ;
  • Oh, Man-Suk (Department of Statisctics, Ewha Womans University)
  • 김민영 (이화여자대학교 통계학과) ;
  • 오만숙 (이화여자대학교 통계학과)
  • Received : 2021.05.30
  • Accepted : 2021.06.16
  • Published : 2021.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

Identification of factors influencing regional population is critical for establishing government's population policies as well as for improving residents' social, economic and cultural well-being in the region. In this study we analysed the data from 2019 Population Housing Survey in Korea to identify the factors affecting the population size in each of the three regions: Seoul, metropolitan cities, and provincial regions. We applied a Bayesian quantile regression to account for asymmetry and heteroscedasticity of data. The analysis results showed that the effects of factors vary greatly between the three regions of Seoul, metropolitan cities, and provincial regions as well as between sub regions within the same region. These results suggest that population-related variables have very heterogeneous characteristics from region to region and therefore it is important to establish customized population policies that suit regional characteristics rather than uniform population policies that apply to every region.

지역별 인구의 분포에 영향을 미치는 요인의 파악은 국가의 사회, 경제, 문화적 발전 위한 정부의 인구정책 수립에 매우 중요하다. 본 연구에서는 2019년 인구주택 총조사 자료를 기반으로 대한민국 국토를 서울, 대도시, 기타지역의 세 지역으로 나누어 각 지역에서 소지역의 인구 크기에 영향을 미치는 요인들을 살펴 보았다. 인구 자료의 특징은 매우 비대칭적이며 이분산성을 가지므로 조건부 평균에 초점을 맞추는 일반적인 회귀모형 대신 분포에 대한 가정이 필요하지 않은 분위회귀모형을 사용하여 인구의 크기에 따라 변화하는 각 요인의 세부적인 영향을 살펴보았다. 분석결과 서울, 대도시, 기타지역에 따라 그리고 같은 지역 내에서도 세부 지역의 인구크기에 따라 요인의 영향이 매우 달라짐을 확인하였다. 이 결과들은 인구관련 변수들이 지역 마다 매우 이질적인 성질을 가지고 있으며 따라서 획일적인 인구정책이 아닌 지역 특성에 맞는 맞춤형 인구정책을 수립해야 할 필요성을 시사한다.

Keywords

Acknowledgement

논문의 향상에 도움이 되는 매우 유용한 코멘트를 주신 두분의 심사위원께 감사를 드린다.

References

  1. Gelman A and Rubin DB (1992). Inference from iterative simulation using multiple sequences, Statistical Science, 7, 457-472. https://doi.org/10.1214/ss/1177011136
  2. Kim BS and Seo WS (2014). Investigating socio-economic characteristics affecting regional population changes : comparing capital region to non-capital region, The Korean Regional Development Association, 26, 1-14.
  3. Kim LY and Seo WS (2016). Investigating spatial patterns and urban influential factors of the school-age population using spatial econometric analysis, The Korean Regional Development Association, 28, 113-129.
  4. Kim LY and Yang KS (2013). Empirical analysis of regional characteristic factors determining net inflow and outflow of the population, The Korean Regional Development Association, 25, 1-19.
  5. Koenker R and Bassett G (1978). Regression quantiles, Econometrica, 46, 33-50. https://doi.org/10.2307/1913643
  6. Lee JH and Jun MJ (2018). An analysis on the redistributive effects of population in the capital region due to the Sejong City construction, The Korean Regional Development Association, 30, 47-65.
  7. Lee YC and Kim CH (2020). A study on the performance evaluation and validation of innovation city development, The Korean Regional Development Association, 32, 47-68.
  8. Min BG (2017). Study on intra-urban migration patterns and the classification of the school-age population in Seoul, The Korean Regional Development Association, 29, 47-72
  9. Oh MS (2019). Bayesian Data Analysis Using JAGS, Freedom Academy Press, Seoul, Korea.
  10. Plummer M (2003). JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling. In Proceedings of the 3rd International Workshop on Distributed Statistical Computing, Technische Universit, Wien, Austria, 125.
  11. Son DG and Hur JW (2018). A study on the effectiveness of the relocating public organization on the reduction of population concentration in the Seoul Metropolitan area, Journal of Korea Planning Association, 53, 5-18.
  12. Yi YJ and Choi MS (2018). Determinants of the elderly's spatio-temporal concentration - using big data of de facto population of Seoul, The Seoul Institute, 19, 149-168.