kavelaltd/historysaid-global-economic

---
license: cc-by-4.0
task_categories:
- tabular-regression
- tabular-classification
tags:
- economics
- macroeconomics
- world-bank
- imf
- time-series
- global-development
size_categories:
- 100K<n<1M
language:
- en
pretty_name: HistorySaid Global Economic Dataset
---
# HistorySaid Global Economic Dataset

[](https://creativecommons.org/licenses/by/4.0/)
[](https://doi.org/10.5281/zenodo.19145374)

Unified economic data from World Bank, IMF, and BIS.
215 countries. 99 indicators. 691,215 observations. One schema.

## Overview

This dataset combines economic, social, and governance indicators from the World Bank (WDI and WGI), IMF World Economic Outlook, and BIS into a single normalized format. It is designed for researchers, data scientists, and developers who need cross-country time-series data without manual harmonization.

## Coverage

| Source | Indicators | Countries | Year Range | Update Frequency |
|---|---|---|---|---|
| BIS | 1 | 106 | 1964–2026 | Annual |
| IMF (WEO) | 3 | 193 | 1980–2024 | Annual |
| World Bank (WDI) | 89 | 215 | 1960–2025 | Annual |
| World Bank (WGI) | 6 | 203 | 1996–2023 | Annual |
| **Total (unified)** | **99** | **215** | **1960–2026** | — |

## Quick Start

### Python

```python
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_parquet("data/unified/all_indicators.parquet")

# GDP for the United States
usa_gdp = df[(df["country_code"] == "USA") & (df["indicator_id"] == "wb.NY.GDP.MKTP.CD")]
usa_gdp = usa_gdp.sort_values("year")

print(usa_gdp[["year", "value"]].tail(10))

plt.figure(figsize=(10, 5))
plt.plot(usa_gdp["year"], usa_gdp["value"] / 1e12)
plt.title("USA GDP (trillions USD)")
plt.xlabel("Year")
plt.ylabel("Trillions USD")
plt.grid(True)
plt.tight_layout()
plt.savefig("usa_gdp.png")
plt.show()
```

### R

```r
library(readr)
library(dplyr)
library(ggplot2)

df <- read_csv("data/unified/all_indicators.csv")

usa_gdp <- df %>%
filter(country_code == "USA", indicator_id == "wb.NY.GDP.MKTP.CD") %>%
arrange(year)

print(tail(usa_gdp %>% select(year, value), 10))

ggplot(usa_gdp, aes(x = year, y = value / 1e12)) +
geom_line() +
labs(title = "USA GDP (trillions USD)", x = "Year", y = "Trillions USD") +
theme_minimal()

ggsave("usa_gdp.png")
```

## Schema

| Column | Type | Description |
|---|---|---|
| country_code | string | ISO 3166-1 alpha-3 code |
| country_name | string | Country name |
| indicator_id | string | Source-namespaced indicator code (e.g., wb.NY.GDP.MKTP.CD) |
| indicator_name | string | Human-readable indicator name |
| source | string | Data source (world_bank, world_bank_wgi, imf_weo, bis) |
| year | integer | Observation year |
| value | float | Observation value (null if not available) |
| unit | string | Unit of measurement |
| scale | string | Scale (always "units") |
| last_updated | string | Date of last data extraction |
| is_estimate | boolean | Whether the value is a projection/estimate |

## File Structure

```
historysaid-global-economic-dataset/
├── README.md
├── LICENSE
├── SOURCES.md
├── METHODOLOGY.md
├── VALIDATION.md
├── CODEBOOK.md
├── CHANGELOG.md
├── .zenodo.json
├── checksums.sha256
├── data/
│ ├── core/
│ │ ├── world_bank.csv / .parquet
│ │ ├── world_bank_wgi.csv / .parquet
│ │ ├── imf_weo.csv / .parquet
│ │ └── bis.csv / .parquet
│ ├── unified/
│ │ ├── all_indicators.csv
│ │ └── all_indicators.parquet
│ └── aggregates/
│ ├── regional_aggregates.csv
│ └── regional_aggregates.parquet
├── _mappings/
│ ├── country_codes.json
│ ├── indicators.json
│ ├── indicator_crosswalk.json
│ ├── coverage_matrix.json
│ └── source_metadata.json
├── _review/ (not included in public release)
│ ├── recon_report.md
│ └── country_code_flags.json
└── examples/
├── quickstart.py
├── quickstart.R
└── sample_queries.md
```

## Sources and Licenses

This dataset redistributes data under the terms of each source's license. See SOURCES.md for full details.

- World Bank (WDI and WGI): CC BY 4.0
- IMF (WEO): IMF Copyright and Usage terms
- BIS: BIS Terms of Use

## Methodology

All country codes are normalized to ISO 3166-1 alpha-3. Indicator codes are namespaced by source (e.g., wb., imf., bis.) to prevent collisions. No interpolation, imputation, or gap-filling was applied. Missing values are null. Full methodology: METHODOLOGY.md

## Validation

Schema compliance: 100% across all sources. Zero duplicate (country_code, indicator_id, year) tuples. Spot-check against source database: 100% match rate. Full report: VALIDATION.md

## Updates

This is version 1.0. Update schedule to be determined.

## Citation

If you use this dataset in published work, please cite:

```bibtex
@dataset{historysaid_global_economic,
author = {{Kavela Ltd}},
title = {HistorySaid Global Economic Dataset},
year = {2025},
publisher = {Zenodo},
version = {1.0},
url = {https://historysaid.com},
doi = {10.5281/zenodo.19145374}
}
```

## Explore

Browse this data interactively at [historysaid.com](https://historysaid.com).