All Issue

2026 Vol.32, Issue 3 Preview Page
Comparison of LSTM-Based Models for Small-Area Household Projections in the Seoul Metropolitan Area
- Applicability of Infrastructure Factors and Spatial Dependency -

수도권 소지역 가구추계를 위한 LSTM 기반 모델 비교
- 인프라 요인과 공간적 의존성의 적용 가능성 -

문소현

Mun, Sohyun

서울대학교 국토문제연구소 연구원
Researcher, The Institute for Korean Regional Studies, Seoul National University
31 August 2026. pp. 33-53
PDF Citation
Abstract
This study examined the applicability of LSTM-based models for small-area future household projections across 66 cities in the Seoul Metropolitan Area (SMA), focusing on differences in predictive performance by input variable configuration. Three models were constructed based on the past trend of the number of households (Model 1), adding infrastructure factors (Model 2), and adding spatial lag factors (Model 3). The evaluation showed that Model 2 generally produced lower MAPE and MdAPE values, but the model with the lowest error differed by region and household type. By household type, Model 2 showed lower errors for one and two person households headed by householders in their 20s or younger and 30s, whereas Model 3 showed lower errors for households headed by householders in their 40s or 50s. These findings suggest that small-area household projections may need to consider region and household-specific input variable configurations, and that future research may extend this approach through ensemble models.
Keywords
household projection
small-area
LSTM model
spatial dependency
local projection model
본 연구는 수도권 66개 시군구를 대상으로 소지역 장래가구추계에서 LSTM 기반 모델의 적용 가능성과 입력 변수 구성에 따른 예측 성능 차이를 검토하였다. 모델은 과거 가구 수 추세 기반(모델 1), 인프라 요인 추가(모델 2), 공간 래그 요인 추가(모델 3) 세 가지로 구축하였다. 성능 평가 결과, 모델 2의 MAPE와 MdAPE가 전반적으로 낮았으나, 지역별·가구 유형별로 낮은 오차를 보이는 모델 구성은 상이하였다. 가구 유형별로는 20대 이하·30대 1·2인 가구에서 모델 2, 40-50대 가구에서 모델 3의 오차가 낮게 나타나 지역 및 가구 특성에 따라 적절한 입력 변수 구성이 달라질 수 있음을 확인하였다. 본 연구는 제한된 소지역 자료 조건 내에서 LSTM 기반 가구추계 모델의 적용 가능성을 검토하고, 지역 및 가구 특성에 따른 앙상블 모델 개발 방향을 제시하였다.
키워드
장래가구추계
소지역
LSTM 모델
공간적 의존성
국지적 추계 모델
References
  1. 국가데이터처, 2022, 「장래가구추계」 통계정보보고서, 국가데이터처.
  2. 김상원·이훈래, 2016, 수도권 도시유형별 내부 인구이동 영향요인 분석, 한국산학기술학회논문지, 17(9), 737-744. 10.5762/KAIS.2016.17.9.737
  3. 문소현·이건학, 2024, 딥러닝 방법을 이용한 소지역 단위 장래가구추계 모델링-다변량 LSTM 모델을 통한 서울시 자치구별 가구추계, 대한지리학회지, 59(6), 790-801.
  4. 유재성, 2021, 가구형태별 주거이동구역 설정: 동적 관점의 커뮤니티 탐지 방법을 이용하여, 국토연구, 111, 45-65.
  5. 이보경·이영주·홍사흠·박미선·오창화, 2021, 주택공급에 따른 인구이동 분석모형 개발 및 활용방안 연구, 국토연구원.
  6. 정재헌·김오석, 2023, 한국 장래인구추계 불확실성 진단: 미래 폭염 노출위험인구 전망을 중심으로, 한국도시지리학회지, 23(6), 69-88.
  7. 조대헌·이상일, 2022, 우리나라 소지역 인구추계를 위한 방법론 비교 분석, 한국지도학회지, 22(2), 51-67. 10.16879/jkca.2022.22.2.051
  8. Ahn, H., Sun, K., and Pio Kim, K., 2022, Comparison of Missing Data Imputation Methods in Time Series Forecasting, Computers, Materials & Continua, 70(1), 767-779. 10.32604/cmc.2022.019369
  9. Alghanmi, N., Alotaibi, R., Alshammari, S., and Mahmood, A., 2024, Population Fusion Transformer for Subnational Population Forecasting, International Journal of Computational Intelligence Systems, 17(1), 26. 10.1007/s44196-024-00413-y
  10. Bryant, J.R. and Graham, P.J., 2013, Bayesian Demographic Accounts: Subnational Population Estimation Using Multiple Data Sources, Bayesian Analysis, 8(3), 591-622. 10.1214/13-BA820
  11. Chi, G. and Voss, P.R., 2011, Small‐area population forecasting: Borrowing strength across space and time, Population, Space and Place, 17(5), 505-520. 10.1002/psp.617
  12. George, M.V., Smith, S.K., Swanson, D.A., and Tayman, J., 2004, Population Projections, in Siegel, J.S. and Swanson, D.A. (eds.), The Methods and Materials of Demography, Elsevier Academic Press, San Diego. 10.1016/B978-012641955-9/50055-2
  13. Gers, F.A., Schmidhuber, J., and Cummins, F., 2000, Learning to forget: Continual prediction with LSTM, Neural Computation, 12(10), 2451-2471. 10.1162/089976600300015015
  14. Glaeser, E.L., Kahn, M.E., and Rappaport, J., 2008, Why do the poor live in cities? The role of public transportation, Journal of Urban Economics, 63(1), 1-24. 10.1016/j.jue.2006.12.004
  15. Grossman, I., Bandara, K., Wilson, T., and Kirley, M., 2022, Can machine learning improve small area population forecasts? A forecast combination approach, Computers, Environment and Urban Systems, 95, 101806. 10.1016/j.compenvurbsys.2022.101806
  16. Grossman, I., Wilson, T., and Temple, J., 2023, Forecasting small area populations with long short-term memory networks, Socio-Economic Planning Sciences, 88, 101658. 10.1016/j.seps.2023.101658
  17. Hamilton, C.H. and Perry, J., 1962, A Short Method for Projecting Population By Age from One Decennial Census to Another, Social Forces, 41(2), 163-170. 10.2307/2573607
  18. Hewamalage, H., Bergmeir, C., and Bandara, K., 2021, Recurrent neural networks for time series forecasting: Current status and future directions, International Journal of Forecasting, 37(1), 388-427. 10.1016/j.ijforecast.2020.06.008
  19. Hochreiter, S. and Schmidhuber, J., 1997, Long short-term memory, Neural Computation, 9(8), 1735-1780. 10.1162/neco.1997.9.8.1735
  20. Hooimeijer, P. and Heida, H., 1994, Small-area household projections: a multi-state demographic design, Nederlandse Geografische Studies, 159-177.
  21. Ip, F. and McRae, D., 1999, Small Area Household Projections - A Parameterised Approach -, Ministry of Finance and Corporate Relations, Province of British Columbia, Canada.
  22. Ma, X., Tao, Z., Wang, Y., Yu, H., and Wang, Y., 2015, Long short-term memory neural network for traffic speed prediction using remote microwave sensor data, Transportation Research Part C: Emerging Technologies, 54, 187-197. 10.1016/j.trc.2015.03.014
  23. Niako, N., 2023, Effects of Missing Data Imputation Methods on Univariate Time Series Forecasting with Arima and LSTM, Master’s thesis, The University of Texas Rio Grande Valley. 10.1186/s12874-024-02448-3 39725886 PMC11670515
  24. Nilsson, I. and Delmelle, E., 2018, Transit investments and neighborhood change: On the likelihood of change, Journal of Transport Geography, 66, 167-179. 10.1016/j.jtrangeo.2017.12.001
  25. Rayer, S. and Smith, S. K., 2010, Factors Affecting the Accuracy of Subcounty Population Forecasts, Journal of Planning Education and Research, 30(2), 147-161. 10.1177/0739456X10380056
  26. Riiman, V., Wilson, A., Milewicz, R., and Pirkelbauer, P., 2019, Comparing Artificial Neural Network and Cohort-Component Models for Population Forecasts, Population Review, 58(2), 100-116. 10.1353/prv.2019.0008
  27. Shi, X., Chen, Z., Wang, H., Yeung, D.-Y., Wong, W., and Woo, W., 2015, Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting, Advances in Neural Information Processing Systems, 28.
  28. Smith, S.K., Tayman, J., and Swanson, D.A., 2013, A Practitioner’s Guide to State and Local Population Projections, Springer, Dordrecht. 10.1007/978-94-007-7551-0
  29. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R., 2014, Dropout: a simple way to prevent neural networks from overfitting, The Journal of Machine Learning Research, 15(1), 1929-1958.
  30. Van Imhoff, E., 1995, Household Demography and Household Modeling, Plenum Press, New York. 10.1007/978-1-4757-5424-7
  31. Wilson, T., 2016, Evaluation of Alternative Cohort-Component Models for Local Area Population Forecasts, Population Research and Policy Review, 35(2), 241-261. 10.1007/s11113-015-9380-y
  32. Wilson, T., Grossman, I., Alexander, M., Rees, P., and Temple, J., 2022, Methods for small area population forecasts: state-of-the-art and research needs, Population Research and Policy Review, 41(3), 865-898. 10.1007/s11113-021-09671-6 34421158 PMC8365292
  33. Wilson, T. and Rowe, F., 2011, The forecast accuracy of local government area population projections: A case study of Queensland, Australasian Journal of Regional Studies, 17(2), 204-243.
  34. KOSIS, 2015·2024, 인구총조사, kosis.kr
  35. 레일포털, data.kric.go.kr
  36. 부동산 114, 아파트 입주 물량, 빅데이터 솔루션(REPS).
Information
  • Publisher :The Korean Association of Regional Geographers
  • Publisher(Ko) :한국지역지리학회
  • Journal Title :REGION AND GEOGRAPHY
  • Journal Title(Ko) :지역과 지리
  • Volume : 32
  • No :3
  • Pages :33-53
  • DOI :https://doi.org/10.26863/JKARG.2026.8.32.3.33
Journal Informaiton REGION AND GEOGRAPHY REGION AND GEOGRAPHY
Journal Informaiton Journal Informaiton - close