load_bolivia

data.load_bolivia()

Load the Bolivia province (ADM2, n=112) PWT-anchored GDP panel.

Subnational GDP for 2012–2022 derived from the 0.25-degree gridded estimates of Rossi-Hansberg & Zhang (2026) under their most aggressive low-population-density censoring (0_05), proportionally rescaled so Bolivian national totals equal Penn World Table 11.0 (rgdpo and pop), and aggregated to GADM 4.10 provinces. GDP and population are therefore in interpretable 2021 PPP US$ units and the relative spatial pattern of the underlying model is preserved exactly.

{citation}

Returns

Name Type Description
gdf geopandas.GeoDataFrame 112 province geometries with columns ["gid", "geometry"], CRS EPSG:4326. Five provinces (BOL.2.1_2, BOL.2.8_2, BOL.2.11_2, BOL.2.13_2, BOL.5.16_2) have no panel rows: all of their grid cells are censored at the 0_05 threshold. geometrics’ alignment warns about them, which is expected.
df pandas.DataFrame Balanced panel of 1177 rows (107 provinces x 11 years, 2012–2022). Key variables: gdp_pwt (millions of 2021 PPP US\(), ``pop_pwt`` (millions of persons), ``gdppc`` (2021 PPP US\) per person) and ln_gdppc, plus provenance/scaling columns documented in the dictionary. Sorted by gid then year.
df_dict pandas.DataFrame Data dictionary with one row per df column, in df column order (gid is the entity, name the entity name, year the time id).

Raises

Name Type Description
GeometricsDataError If a source file cannot be downloaded or fails hash verification.

See Also

load_bolivia_departments : Department-level (ADM1, n=9) version. load_bolivia_grid : The underlying 0.25-degree grid cells (n=1603). load_bolivia_raw : Untouched files of any level, including ADM0.

Examples

>>> from geometrics.data import load_bolivia
>>> gdf, df, df_dict = load_bolivia()
>>> df.shape
(1177, 21)