electricsheepafrica/africa-south-sudan-real-time-prices

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- 1K<n<10K
source_datasets:
- original
task_categories:
- tabular-regression
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- energy
- food-security
- ssd
pretty_name: "South Sudan - Real Time Prices"
dataset_info:
splits:
- name: train
num_examples: 7207
- name: test
num_examples: 1801
---

# South Sudan - Real Time Prices

**Publisher:** World Bank Group · **Source:** [HDX](https://data.humdata.org/dataset/south-sudan-real-time-prices) · **License:** `cc-by` · **Updated:** 2026-04-01

---

## Abstract

Real Time Prices (RTP) is a live dataset compiled and updated weekly by the World Bank Development Economics Data Group (DECDG) using a combination of direct price measurement and Machine Learning estimation of missing price data. The historical and current estimates are based on price information gathered from the World Food Program (WFP), UN-Food and Agricultural Organization (FAO), select National Statistical Offices, and are continually updated and revised as more price information becomes available. Real-time exchange rate data used in this process are from official and public sources.

RTP includes three sub-series, Real Time Food Prices (RTFP) includes prices on a variety of food items that primarily include country-specific staple foods, Real Time Energy Prices (RTEP) includes fuel prices, and Real Time Exchange Rates (RTFX) and includes unofficial exchange rate estimates as well as possible other unofficial deflators.

Each row in this dataset represents country-level aggregates. Temporal coverage is indicated by the `dates`, `start_dense_data` column(s). Geographic scope: **SSD**.

*Curated into ML-ready Parquet format by [Electric Sheep Africa](https://huggingface.co/electricsheepafrica).*

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Food security and nutrition |
| **Unit of observation** | Country-level aggregates |
| **Rows (total)** | 9,009 |
| **Columns** | 107 (94 numeric, 10 categorical, 3 datetime) |
| **Train split** | 7,207 rows |
| **Test split** | 1,801 rows |
| **Geographic scope** | SSD |
| **Publisher** | World Bank Group |
| **HDX last updated** | 2026-04-01 |

---

## Variables

**Geographic** — `iso3` (SSD), `country` (South Sudan), `lat` (range 4.24–11.74), `lon` (range 26.87–33.73), `year` (range 2007.0–2026.0) and 23 others.

**Temporal** — `dates`, `month` (range 1.0–12.0).

**Demographic** — `data_coverage` (range 27.12–27.12), `data_coverage_recent` (range 42.44–42.44), `o_livestockgoat_male`, `h_livestockgoat_male`, `l_livestockgoat_male` and 7 others.

**Identifier / Metadata** — `adm1_name` (Unity, Jonglei, Warrap), `adm2_name` (Pariang, Twic, Juba), `mkt_name` (Abyei, Suk Shabi, Old Fangak), `geo_id` (gid_93500000282800000, gid_117400000328100000, gid_90700000308800000), `esa_source` (HDX) and 1 others.

**Other** — `components` (beans (1 KG, Index Weight = 1), fuel_diesel (1 L, Index Weight = 0), fuel_petrol_gasoline (1 L, Index Weight = 0), fuel_petrol_gasoline_parallel_market (1 L, Index Weight = 0), groundnuts (1 KG, Index Weight = 1), livestockgoat_male (1 Head, Index Weight = 0), livestocksheep_male (1 Head, Index Weight = 0), maize (3.5 KG, Index Weight = 0.29), maize_meal (1 KG, Index Weight = 1), millet (3.5 KG, Index Weight = 0.29), milling_cost_sorghum (LCU/3.5kg, Index Weight = 0), oil (1 L, Index Weight = 1), rice (1 KG, Index Weight = 1), salt (1 KG, Index Weight = 1), sesame (3.5 KG, Index Weight = 0.29), sorghum (3.5 KG, Index Weight = 0.29), wage_non_qualified_labour_non_agricultural (1 Day, Index Weight = 0), wheat_flour (1 KG, Index Weight = 1)), `start_dense_data`, `beans` (range -8.79–71100.0), `sorghum` (range -54.39–70000.0), `o_beans` (range 3.92–47630.83) and 54 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-south-sudan-real-time-prices")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

print(train.shape)
train.head()
```

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `iso3` | object | 0.0% | SSD |
| `country` | object | 0.0% | South Sudan |
| `adm1_name` | object | 0.0% | Unity, Jonglei, Warrap |
| `adm2_name` | object | 0.0% | Pariang, Twic, Juba |
| `mkt_name` | object | 0.0% | Abyei, Suk Shabi, Old Fangak |
| `lat` | float64 | 2.6% | 4.24 – 11.74 (mean 8.0134) |
| `lon` | float64 | 2.6% | 26.87 – 33.73 (mean 30.2797) |
| `geo_id` | object | 0.0% | gid_93500000282800000, gid_117400000328100000, gid_90700000308800000 |
| `dates` | datetime64[ns] | 0.0% | |
| `year` | int64 | 0.0% | 2007.0 – 2026.0 (mean 2016.1299) |
| `month` | int64 | 0.0% | 1.0 – 12.0 (mean 6.4416) |
| `currency` | object | 0.0% | SSP |
| `components` | object | 0.0% | beans (1 KG, Index Weight = 1), fuel_diesel (1 L, Index Weight = 0), fuel_petrol_gasoline (1 L, Index Weight = 0), fuel_petrol_gasoline_parallel_market (1 L, Index Weight = 0), groundnuts (1 KG, Index Weight = 1), livestockgoat_male (1 Head, Index Weight = 0), livestocksheep_male (1 Head, Index Weight = 0), maize (3.5 KG, Index Weight = 0.29), maize_meal (1 KG, Index Weight = 1), millet (3.5 KG, Index Weight = 0.29), milling_cost_sorghum (LCU/3.5kg, Index Weight = 0), oil (1 L, Index Weight = 1), rice (1 KG, Index Weight = 1), salt (1 KG, Index Weight = 1), sesame (3.5 KG, Index Weight = 0.29), sorghum (3.5 KG, Index Weight = 0.29), wage_non_qualified_labour_non_agricultural (1 Day, Index Weight = 0), wheat_flour (1 KG, Index Weight = 1) |
| `start_dense_data` | datetime64[ns] | 0.0% | |
| `last_survey_point` | datetime64[ns] | 0.0% | |
| `data_coverage` | float64 | 0.0% | 27.12 – 27.12 (mean 27.12) |
| `data_coverage_recent` | float64 | 0.0% | 42.44 – 42.44 (mean 42.44) |
| `index_confidence_score` | float64 | 0.0% | 0.96 – 0.96 (mean 0.96) |
| `spatially_interpolated` | int64 | 0.0% | 0.0 – 0.0 (mean 0.0) |
| `beans` | float64 | 77.7% | -8.79 – 71100.0 (mean 2657.351) |
| `sorghum` | float64 | 77.2% | -54.39 – 70000.0 (mean 3162.6668) |
| `o_beans` | float64 | 0.0% | 3.92 – 47630.83 (mean 1593.7026) |
| `h_beans` | float64 | 0.0% | 4.22 – 52881.05 (mean 1771.2352) |
| `l_beans` | float64 | 0.0% | 3.57 – 25925.05 (mean 1416.8487) |
| `c_beans` | float64 | 0.0% | 4.0 – 41722.04 (mean 1565.069) |
| `inflation_beans` | float64 | 5.2% | -68.92 – 1236.8 (mean 74.9386) |
| `trust_beans` | float64 | 0.0% | 8.9 – 10.0 (mean 9.1566) |
| `o_groundnuts` | float64 | 0.0% | 2.8 – 21351.28 (mean 1167.0635) |
| `h_groundnuts` | float64 | 0.0% | 3.07 – 22507.76 (mean 1275.8031) |
| `l_groundnuts` | float64 | 0.0% | 2.69 – 19503.89 (mean 1062.9093) |
| `c_groundnuts` | float64 | 0.0% | 2.88 – 20275.1 (mean 1158.7543) |
| `inflation_groundnuts` | float64 | 5.2% | |
| `trust_groundnuts` | float64 | 0.0% | |
| `o_livestockgoat_male` | float64 | 0.0% | |
| `h_livestockgoat_male` | float64 | 0.0% | |
| `l_livestockgoat_male` | float64 | 0.0% | |
| `c_livestockgoat_male` | float64 | 0.0% | |
| `inflation_livestockgoat_male` | float64 | 5.2% | |
| `trust_livestockgoat_male` | float64 | 0.0% | |
| `o_livestocksheep_male` | float64 | 0.0% | |
| `h_livestocksheep_male` | float64 | 0.0% | |
| `l_livestocksheep_male` | float64 | 0.0% | |
| `c_livestocksheep_male` | float64 | 0.0% | |
| `inflation_livestocksheep_male` | float64 | 5.2% | |
| `trust_livestocksheep_male` | float64 | 0.0% | |
| `o_maize` | float64 | 0.0% | |
| `h_maize` | float64 | 0.0% | |
| `l_maize` | float64 | 0.0% | |
| `c_maize` | float64 | 0.0% | |
| `inflation_maize` | float64 | 5.2% | |
| `trust_maize` | float64 | 0.0% | |
| `o_maize_meal` | float64 | 0.0% | |
| `h_maize_meal` | float64 | 0.0% | |
| `l_maize_meal` | float64 | 0.0% | |
| `c_maize_meal` | float64 | 0.0% | |
| `inflation_maize_meal` | float64 | 5.2% | |
| `trust_maize_meal` | float64 | 0.0% | |
| `o_millet` | float64 | 0.0% | |
| `h_millet` | float64 | 0.0% | |
| `l_millet` | float64 | 0.0% | |
| `c_millet` | float64 | 0.0% | |
| `inflation_millet` | float64 | 5.2% | |
| `trust_millet` | float64 | 0.0% | |
| `o_oil` | float64 | 0.0% | |
| `h_oil` | float64 | 0.0% | |
| `l_oil` | float64 | 0.0% | |
| `c_oil` | float64 | 0.0% | |
| `inflation_oil` | float64 | 5.2% | |
| `trust_oil` | float64 | 0.0% | |
| `o_rice` | float64 | 0.0% | |
| `h_rice` | float64 | 0.0% | |
| `l_rice` | float64 | 0.0% | |
| `c_rice` | float64 | 0.0% | |
| `inflation_rice` | float64 | 5.2% | |
| `trust_rice` | float64 | 0.0% | |
| `o_salt` | float64 | 0.0% | |
| `h_salt` | float64 | 0.0% | |
| `l_salt` | float64 | 0.0% | |
| `c_salt` | float64 | 0.0% | |
| `inflation_salt` | float64 | 5.2% | |
| `trust_salt` | float64 | 0.0% | |
| `o_sesame` | float64 | 0.0% | |
| `h_sesame` | float64 | 0.0% | |
| `l_sesame` | float64 | 0.0% | |
| `c_sesame` | float64 | 0.0% | |
| `inflation_sesame` | float64 | 5.2% | |
| `trust_sesame` | float64 | 0.0% | |
| `o_sorghum` | float64 | 0.0% | |
| `h_sorghum` | float64 | 0.0% | |
| `l_sorghum` | float64 | 0.0% | |
| `c_sorghum` | float64 | 0.0% | |
| `inflation_sorghum` | float64 | 5.2% | |
| `trust_sorghum` | float64 | 0.0% | |
| `o_wheat_flour` | float64 | 0.0% | |
| `h_wheat_flour` | float64 | 0.0% | |
| `l_wheat_flour` | float64 | 0.0% | |
| `c_wheat_flour` | float64 | 0.0% | |
| `inflation_wheat_flour` | float64 | 5.2% | |
| `trust_wheat_flour` | float64 | 0.0% | |
| `o_food_price_index` | float64 | 0.0% | |
| `h_food_price_index` | float64 | 0.0% | |
| `l_food_price_index` | float64 | 0.0% | |
| `c_food_price_index` | float64 | 0.0% | |
| `inflation_food_price_index` | float64 | 5.2% | |
| `trust_food_price_index` | float64 | 0.0% | |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-08 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `lat` | 4.24 | 11.74 | 8.0134 | 8.67 |
| `lon` | 26.87 | 33.73 | 30.2797 | 30.125 |
| `year` | 2007.0 | 2026.0 | 2016.1299 | 2016.0 |
| `month` | 1.0 | 12.0 | 6.4416 | 6.0 |
| `data_coverage` | 27.12 | 27.12 | 27.12 | 27.12 |
| `data_coverage_recent` | 42.44 | 42.44 | 42.44 | 42.44 |
| `index_confidence_score` | 0.96 | 0.96 | 0.96 | 0.96 |
| `spatially_interpolated` | 0.0 | 0.0 | 0.0 | 0.0 |
| `beans` | -8.79 | 71100.0 | 2657.351 | 675.0 |
| `sorghum` | -54.39 | 70000.0 | 3162.6668 | 638.0 |
| `o_beans` | 3.92 | 47630.83 | 1593.7026 | 67.74 |
| `h_beans` | 4.22 | 52881.05 | 1771.2352 | 78.55 |
| `l_beans` | 3.57 | 25925.05 | 1416.8487 | 60.0 |
| `c_beans` | 4.0 | 41722.04 | 1565.069 | 75.64 |
| `inflation_beans` | -68.92 | 1236.8 | 74.9386 | 36.19 |

---

## Curation

Raw data was downloaded from HDX via the CKAN API and converted to Parquet. Column names were lowercased and standardised to snake_case. Common missing-value markers (`N/A`, `null`, `none`, `-`, `unknown`, `no data`, `#N/A`) were unified to `NaN`. 753 column(s) with >80% missing values were removed: `apples`, `bananas`, `beans_egyptian`, `beans_fao`, `bread`, `bread_fao`.... 3 column(s) were cast from string to numeric or datetime based on parse-success rate (>85% threshold). The dataset was split 80/20 into train and test partitions using a fixed random seed (42) and saved as Snappy-compressed Parquet.

---

## Limitations

- Data originates from World Bank Group and has not been independently validated by ESA.
- Automated cleaning cannot correct for misreported values, definitional inconsistencies, or sampling bias in the original collection.
- The following columns have >20% missing values and should be treated with caution in modelling: `beans`, `sorghum`.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/south-sudan-real-time-prices) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_south_sudan_real_time_prices,
title = {South Sudan - Real Time Prices},
author = {World Bank Group},
year = {2026},
url = {https://data.humdata.org/dataset/south-sudan-real-time-prices},
note = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}
```

---

*[Electric Sheep Africa](https://huggingface.co/electricsheepafrica) — Africa's ML dataset infrastructure. Lagos, Nigeria.*