Multi-Source County-Year Dataset for Tertiary Industry Added Value Reconstruction in Shanx

A processed county-year panel dataset for reconstructing tertiary industry added value in Shanxi Province, China, from 2012 to 2022. It integrates multi-source remote sensing, geospatial, and socioeconomic information within a unified framework. The dataset was created by PENGFEI Jia and includes processed predictor tables, metadata, and reproducible code files.

Use Cases

• Reconstructing county-level economic output based on integrated remote sensing and statistical data.
• Performing feature selection and interpretability analysis using the provided feature subsets and SHAP analysis code.
• Assessing spatial consistency of economic indicators at the county level across an 11-year period.
• Training machine learning models to predict tertiary industry added value from multi-source geospatial predictors.

Strengths

• Integrates data from over 10 distinct sources, including Landsat, MODIS, and official statistical yearbooks.
• Covers an 11-year time range (2012–2022) for Shanxi Province.
• Includes reproducible code for model fitting, SHAP analysis, and spatial evaluation.
• Released under a permissive CC-BY-4.0 license.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset shares processed analytical data rather than the original raw third-party source data.

Provenance

Source

figshare, author PENGFEI Jia.

Collection Method

Processed and integrated from multi-source remote sensing, open geospatial, and socioeconomic data.

Time Range

2012–2022

Freshness

Last updated 2026-05-09 13:33:33; freshness should be verified.

Geography

Shanxi Province, China, at the county level.