electricsheepafrica/african-financial-survey-database-2016

---
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
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- economics
- gross-domestic-product-gdp
- dza
- ago
- ben
- bwa
- bfa
pretty_name: "African Financial Survey Database, 2016"
dataset_info:
splits:
- name: train
num_examples: 31104
- name: test
num_examples: 7776
---

# African Financial Survey Database, 2016

**Publisher:** African Development Bank Group · **Source:** [HDX](https://data.humdata.org/dataset/african-financial-survey-database-2016) · **License:** `cc-by` · **Updated:** 2023-03-03

---

## Abstract

African Financial Survey Database, 2005-2014

Each row in this dataset represents first-level administrative unit observations. Data was last updated on HDX on 2023-03-03. Geographic scope: **DZA, AGO, BEN, BWA, BFA, BDI, CPV, CMR, and 50 others**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Humanitarian and development data |
| **Unit of observation** | First-level administrative unit observations |
| **Rows (total)** | 38,880 |
| **Columns** | 15 (2 numeric, 13 categorical, 0 datetime) |
| **Train split** | 31,104 rows |
| **Test split** | 7,776 rows |
| **Geographic scope** | DZA, AGO, BEN, BWA, BFA, BDI, CPV, CMR, and 50 others |
| **Publisher** | African Development Bank Group |
| **HDX last updated** | 2023-03-03 |

---

## Variables

**Geographic** — `countryname` (Algeria, Senegal, Malawi), `pays` (Algérie, Sénégal, Malawi), `country` (DZA, SEN, MWI), `currency` (CFA Franc, Algerian Dinar, Somali Shilling), `region` (West Africa, East Africa, Southern Africa).

**Temporal** — `date` (range 2005.0–2014.0).

**Outcome / Measurement** — `value` (range -16.2124–12686454.0197).

**Identifier / Metadata** — `indicatorname` (Total Value of RTGS Transactions, Total Population - Million, Assets/ Capita), `esa_source`, `esa_processed`.

**Other** — `capital` (Algiers, Dakar, Lilongwe), `capitale` (Alger, Dakar, Lilongwe), `monnaie` (Franc CFA, Kwacha, Dinar Algérien), `indicator` (AF1, AF2, AF61), `unit`.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/african-financial-survey-database-2016")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `countryname` | object | 0.0% | Algeria, Senegal, Malawi |
| `pays` | object | 0.0% | Algérie, Sénégal, Malawi |
| `country` | object | 0.0% | DZA, SEN, MWI |
| `capital` | object | 0.0% | Algiers, Dakar, Lilongwe |
| `capitale` | object | 0.0% | Alger, Dakar, Lilongwe |
| `currency` | object | 0.0% | CFA Franc, Algerian Dinar, Somali Shilling |
| `monnaie` | object | 0.0% | Franc CFA, Kwacha, Dinar Algérien |
| `region` | object | 0.0% | West Africa, East Africa, Southern Africa |
| `indicator` | object | 0.0% | AF1, AF2, AF61 |
| `indicatorname` | object | 0.0% | Total Value of RTGS Transactions, Total Population - Million, Assets/ Capita |
| `date` | int64 | 0.0% | 2005.0 – 2014.0 (mean 2009.5) |
| `unit` | object | 4.2% | |
| `value` | float64 | 28.6% | -16.2124 – 12686454.0197 (mean 22450.4383) |
| `esa_source` | object | 0.0% | |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `date` | 2005.0 | 2014.0 | 2009.5 | 2009.5 |
| `value` | -16.2124 | 12686454.0197 | 22450.4383 | 36.3845 |

---

## 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`. 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 African Development 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: `value`.
- This dataset spans 58 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/african-financial-survey-database-2016) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_african_financial_survey_database_2016,
title = {African Financial Survey Database, 2016},
author = {African Development Bank Group},
year = {2023},
url = {https://data.humdata.org/dataset/african-financial-survey-database-2016},
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.*