Transfer Learning for Socioeconomic Estimation in Forced-Displacement Settings

• URL: http://arxiv.org/abs/2609.15773v1
• Date: Mon, 14 Sep 2026 15:55:57 GMT
• Title: Transfer Learning for Socioeconomic Estimation in Forced-Displacement Settings
• Abstract summary: More frequently updated, spatially complementary evidence is needed to identify where socioeconomic conditions may be changing between survey rounds.<n>We develop and evaluate an updated, adapted model using socioeconomic indices derived from UNHCR FDS and RMS data.<n>These estimates sustain insight between survey rounds and support timely humanitarian prioritization and field verification.
• Score: 1.7804100659374038
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Progress in inclusive household surveys has strengthened socioeconomic evidence for forcibly displaced populations, providing indispensable benchmarks on living conditions and welfare. However, these surveys remain resource-intensive and periodic, while conditions can change between rounds, particularly in settings affected by fragility, conflict, and violence. More frequently updated, spatially granular complementary evidence is therefore needed to identify where socioeconomic conditions may be changing between survey rounds and to inform operational prioritization. Earth observation and machine learning offer a scalable source of spatially explicit socioeconomic information. However, tools developed for general populations have not been systematically adapted and evaluated in forced displacement settings, where living conditions, settlement patterns, and displacement impacts may differ substantially. We address this gap by adapting a multimodal spatiotemporal vision transformer, pretrained on Demographic and Health Survey data from approximately 1.2 million households across 36 African countries, to forced displacement and host community settings in South Sudan, Cameroon, and Zambia. We develop and evaluate the updated, adapted model using socioeconomic indices derived from UNHCR FDS and RMS data. Our results show that satellite-derived geospatial covariates explain up to 66% of the variation in socioeconomic outcomes in camp-intersecting grids, with a mean absolute error (MAE) of 4.37 index points, and 41% in non-camp-intersecting areas, with an MAE of 5.41. The framework complements and adds value to periodic household surveys by filling critical spatial and temporal data gaps with regularly updated, model-based socioeconomic estimates. These estimates sustain insight between survey rounds and support timely humanitarian prioritization and field verification.

Related papers

• Geospatial Foundation Models Capture Health-Relevant Dimensions of Place Beyond Conventional Social Risk Indices [1.233122988113145]
  Geospatial foundation models capture health-relevant features of place not represented by conventional social risk indices.<n>For health outcomes, models explained up to 54% of variance left unexplained by social risk indices.
  arXiv  Detail & Related papers  (2026-09-10T15:16:15Z)
• Joint Utilization of Geospatial and census proxies for Autoencoder-Assisted Downscaling (JUGAAD) of socioeconomic indicators in India [6.076588686970015]
  This study introduces a deep learning framework, JuGAAD, using Indian census and survey data from 2001 and 2011 as a case study.<n>We employ a three-step process: census and geospatial data are averaged into intermediate village-cluster-scale tessellations to reduce noise and regularize administrative boundary changes.<n>An autoencoder compresses high-dimensional National Sample Survey Office (NSSO) data into a low-dimensional latent representation; and a regression model maps upscaled census and geospatial data to this representation.<n>This function is applied to fine-grained census data to generate high-resolution predictions, validated against ground-
  arXiv  Detail & Related papers  (2026-07-17T21:57:55Z)
• Uncovering Salience-Driven Dynamics in Consumer Confidence with Generative Social Simulation [56.70690339492564]
  We introduce ConsumerSim, a generative Human--Environment response framework that reconstructs Consumer Confidence Index dynamics.<n>Across U.S., EU27, and Japanese official CCI target series, ConsumerSim ranks first among persistence, time-series, regression, and shocks information-augmented baselines.<n>Findings support a behavioral view of consumer confidence as an interpretable Human--Environment response process.
  arXiv  Detail & Related papers  (2026-06-29T14:46:22Z)
• Benchmarking LLMs for Community Governance Simulation with Life-history Narratives [46.86050402684712]
  Large language models (LLMs) are scalable to simulate human attitudes and behaviors at low cost.<n>We present an integrated research framework covering dataset, benchmark, algorithm, and system.<n>System integrates curriculum-LoRA into a closed-loop policy-evaluation pipeline.
  arXiv  Detail & Related papers  (2026-05-22T15:48:49Z)
• Geospatial foundation-model embeddings improve population estimation unevenly across space and scale [4.055687405756816]
  Population Dynamics Foundation Model (PDFM) was tested in Brazil, Nigeria and the United States.<n>PDFM increased predictive fit by a median of 20.1% (IQR: 10.0-33.2%) reduction in unexplained variance, and reduced Kullback-Leibler divergence by 23.2% (9.2-26.2%)
  arXiv  Detail & Related papers  (2026-05-03T00:20:08Z)
• A satellite foundation model for improved wealth monitoring [45.63770015866594]
  Poverty statistics guide social policy, but in many low- and middle-income countries, censuses and household surveys that collect these data are costly, infrequent, quickly outdated, and sometimes error-prone.<n>Here we introduce Tempov, a satellite foundation model pretrained by self-supervision on three million bi-temporal Landsat pairs and adapted with parameter-efficient fine-tuning to sparse survey labels.<n>The model enables large-scale, high-resolution wealth mapping and dynamic measurement, including zero-shot nowcasting up to a decade after observed labels, retrospective hindcasting, and decadal change tracking.
  arXiv  Detail & Related papers  (2026-04-25T06:37:51Z)
• Time Profile of U.S. Neighborhoods: Datasets of Time Use at Social Infrastructure Places [5.244716758485668]
  Social infrastructure plays a critical role in shaping neighborhood well-being by fostering social interaction, enabling service provision, and encouraging exposure to diverse environments.<n>Despite the growing knowledge of its spatial accessibility, time use at social infrastructure places is underexplored due to the lack of a spatially resolved national dataset.<n>We develop scalable Social-Infrastructure Time Use measures (STU) that capture length and depth of engagement, activity diversity, and spatial inequality.<n>Our datasets leverage anonymized and aggregated foot traffic data collected between 2019 and 2024 across 49 continental U.S. states.
  arXiv  Detail & Related papers  (2025-08-18T18:22:04Z)
• Comparative Analysis of Time Series Foundation Models for Demographic Forecasting: Enhancing Predictive Accuracy in US Population Dynamics [4.3009067533625895]
  This study explores the application of time series foundation models to predict demographic changes in the United States.<n>We evaluate the performance of the Time Series Foundation Model (TimesFM) against traditional baselines.<n>Our experiments across six demographically diverse states demonstrate that TimesFM achieves the lowest Mean Squared Error (MSE) in 86.67% of test cases.
  arXiv  Detail & Related papers  (2025-08-09T07:55:09Z)
• Space-aware Socioeconomic Indicator Inference with Heterogeneous Graphs [46.52719756897067]
  We present GeoHG, the first space-aware socioeconomic indicator inference method that utilizes a heterogeneous graph-based structure to represent geospace for non-continuous inference.
  arXiv  Detail & Related papers  (2024-05-23T03:19:02Z)
• Magnify Your Population: Statistical Downscaling to Augment the Spatial Resolution of Socioeconomic Census Data [48.7576911714538]
  We present a new statistical downscaling approach to derive fine-scale estimates of key socioeconomic attributes. For each selected socioeconomic variable, a Random Forest model is trained on the source Census units and then used to generate fine-scale gridded predictions. As a case study, we apply this method to Census data in the United States, downscaling the selected socioeconomic variables available at the block group level, to a grid of 300 spatial resolution.
  arXiv  Detail & Related papers  (2020-06-23T16:52:18Z)
• Entropy as a measure of attractiveness and socioeconomic complexity in Rio de Janeiro metropolitan area [52.77024349608834]
  We use a mobile phone dataset and an entropy-based metric to measure the attractiveness of a location. The results show that the attractiveness of a given location measured by entropy is an important descriptor of the socioeconomic status of the location.
  arXiv  Detail & Related papers  (2020-03-23T15:58:56Z)
• Survival Cluster Analysis [93.50540270973927]
  There is an unmet need in survival analysis for identifying subpopulations with distinct risk profiles. An approach that addresses this need is likely to improve characterization of individual outcomes.
  arXiv  Detail & Related papers  (2020-02-29T22:41:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.