electricsheepafrica/africa-global-market-monitor

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- 10K<n<100K
source_datasets:
- original
task_categories:
- tabular-regression
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- crisis-greater-middle-east
- markets
- afg
- dza
- ago
- arm
- bgd
pretty_name: "WFP Global Market Monitor"
dataset_info:
splits:
- name: train
num_examples: 9565
- name: test
num_examples: 2391
---

# WFP Global Market Monitor

**Publisher:** WFP - World Food Programme · **Source:** [HDX](https://data.humdata.org/dataset/global-market-monitor) · **License:** `cc-by-igo` · **Updated:** 2026-03-31

---

## Abstract

The [WFP Global Market Monitor](https://www.wfp.org/publications/market-monitor) monitors food prices in markets across a range of countries globally. The resources are updated every other week, with country data updating based on when data is received from country offices.

Each row in this dataset represents first-level administrative unit observations. Temporal coverage is indicated by the `date`, `lastmodifydate` column(s). Geographic scope: **AFG, DZA, AGO, ARM, BGD, BEN, BOL, BFA, and 72 others**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Food security and nutrition |
| **Unit of observation** | First-level administrative unit observations |
| **Rows (total)** | 11,957 |
| **Columns** | 25 (11 numeric, 12 categorical, 2 datetime) |
| **Train split** | 9,565 rows |
| **Test split** | 2,391 rows |
| **Geographic scope** | AFG, DZA, AGO, ARM, BGD, BEN, BOL, BFA, and 72 others |
| **Publisher** | WFP - World Food Programme |
| **HDX last updated** | 2026-03-31 |

---

## Variables

**Geographic** — `monthlyversion` (range 50.0–113.0), `frequencyname` (Monthly), `countrycode` (range 1.0–70001.0), `countryname` (Mauritania, Rwanda, Pakistan), `admin1` (National Average , National Average ) and 12 others.

**Temporal** — `date`, `pricetrendquarter` (Negative, Moderate, Normal ), `pricetrendmonth` (N/A , Negative, Normal ).

**Identifier / Metadata** — `esa_source`, `esa_processed`.

**Other** — `datalevel` (National ), `mainstaplefood` (Sugar , Wheat flour , Rice (imported) ), `caloriccontribution` (range 5.0–70.0).

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-global-market-monitor")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `date` | datetime64[ns] | 0.0% | |
| `monthlyversion` | int64 | 0.0% | 50.0 – 113.0 (mean 78.5272) |
| `frequencyname` | object | 0.0% | Monthly |
| `countrycode` | int64 | 0.0% | 1.0 – 70001.0 (mean 2181.7929) |
| `countryname` | object | 0.0% | Mauritania, Rwanda, Pakistan |
| `admin1` | object | 0.0% | National Average , National Average |
| `datalevel` | object | 0.0% | National |
| `mainstaplefood` | object | 0.0% | Sugar , Wheat flour , Rice (imported) |
| `pricetype` | object | 0.0% | Retail , Wholesale |
| `caloriccontribution` | float64 | 0.2% | 5.0 – 70.0 (mean 17.332) |
| `quarterlychangensa` | float64 | 14.9% | -100.0 – 658.0 (mean 4.3069) |
| `monthlychangensa` | float64 | 45.0% | -74.0 – 193.0 (mean 1.5485) |
| `quarterlychangesa` | float64 | 15.1% | -100.0 – 847.0 (mean 2.5239) |
| `monthlychangesa` | float64 | 45.2% | -74.0 – 508.0 (mean 0.9105) |
| `yoychangequarter` | float64 | 15.5% | -100.0 – 1942.0 (mean 21.5421) |
| `yoychangemonth` | float64 | 15.8% | -100.0 – 1796.0 (mean 22.3996) |
| `pricetrendquarter` | object | 0.0% | Negative, Moderate, Normal |
| `pricetrendmonth` | object | 0.0% | N/A , Negative, Normal |
| `quarterlycostshare` | float64 | 14.8% | 2.0 – 100.0 (mean 32.5415) |
| `totimpactquarterlychange` | float64 | 72.4% | -100.0 – 296.0 (mean 1.709) |
| `totimpactquarterlycode` | object | 0.0% | Negative, Moderate, Normal |
| `totimpactmonthlycode` | object | 0.0% | N/A , Negative, Normal |
| `lastmodifydate` | datetime64[ns] | 0.0% | |
| `esa_source` | object | 0.0% | |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `monthlyversion` | 50.0 | 113.0 | 78.5272 | 76.0 |
| `countrycode` | 1.0 | 70001.0 | 2181.7929 | 159.0 |
| `caloriccontribution` | 5.0 | 70.0 | 17.332 | 12.0 |
| `quarterlychangensa` | -100.0 | 658.0 | 4.3069 | 1.0 |
| `monthlychangensa` | -74.0 | 193.0 | 1.5485 | 0.0 |
| `quarterlychangesa` | -100.0 | 847.0 | 2.5239 | 1.0 |
| `monthlychangesa` | -74.0 | 508.0 | 0.9105 | 0.0 |
| `yoychangequarter` | -100.0 | 1942.0 | 21.5421 | 9.0 |
| `yoychangemonth` | -100.0 | 1796.0 | 22.3996 | 9.0 |
| `quarterlycostshare` | 2.0 | 100.0 | 32.5415 | 24.0 |
| `totimpactquarterlychange` | -100.0 | 296.0 | 1.709 | 1.0 |

---

## 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`. 1 column(s) with >80% missing values were removed: `totimpactmonthlychange`. 2 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 WFP - World Food Programme 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: `monthlychangensa`, `monthlychangesa`, `totimpactquarterlychange`.
- This dataset spans 80 countries; geographic and methodological inconsistencies across national boundaries may affect cross-country comparability.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/global-market-monitor) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_global_market_monitor,
title = {WFP Global Market Monitor},
author = {WFP - World Food Programme},
year = {2026},
url = {https://data.humdata.org/dataset/global-market-monitor},
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.*