Humanity Forecast
← Back

Data & Methodology

All data, code, and results used in this project are freely available for download, verification, and reuse. We believe in full transparency — every number on this site can be traced back to these source files.

Reproducibility Verified

100 TFR scenarios (0.1 to 10.0) computed with 10M trajectories each — 1 billion total trajectories on 8× NVIDIA A100 GPUs. Original 6 named scenarios verified with 5 independent runs each. Maximum coefficient of variation: 2.55%. All other data points: 0.00% CV.

Simulation Results

tfr_sweep_global.json(443 KB)

Complete TFR sweep — 100 scenarios (TFR 0.1 to 10.0 in 0.1 steps) × 10M trajectories each = 1 billion total. Contains p5/p25/p50/p75/p95 percentiles and the probability of falling below 1M people at every 5-year step from 2029-2499. The public hero uses the most interpretable 0.5-5.0 range; extreme high-TFR values above 5.0 in the raw file should be treated as illustrative stress tests.

Download
scenario_results_verified.json(3.8 KB)

Original 6 named scenario projections (1M trajectories each). Contains timeline data points and threshold crossing dates (e.g. year population drops below 1B).

Download
reproducibility_test.json(16 KB)

Reproducibility analysis — each scenario run 5 independent times with different random seeds. Reports coefficient of variation (CV) at each time point. Maximum CV: 2.55% at one outlier point due to histogram bin resolution; all others 0.00%.

Download
country_projections.csv(12 KB)

Per-country population projections (192 sovereign nations) at years 2024-2300 under UN medium variant. CSV format, values in millions.

Download
country_projections_full.json(38 KB)

Extended country data including language, religion, and population projections. JSON format.

Download

Per-Country Data

countries/{slug}.json(~350 KB each)

Per-country TFR sweep — 100 TFR scenarios (0.1-10.0) with p5/p25/p50/p75/p95 percentiles at every 5-year step. 192 files, one per country. Powers the interactive TFR slider on country pages.

Download
economic/{slug}.json(~120 KB each)

Per-country economic projections — dependency ratios, GDP (Barro convergence model), fiscal pressure index, and working-age share under 13 TFR scenarios. 192 files. These are derived estimates, not GPU-simulated.

Download

Source Data (Third Party)

un_wpp_2024_demographics.csv(3.9 MB)

UN World Population Prospects 2024 — demographic indicators for 237 countries (2024-2101). Includes TFR, life expectancy, CBR, CDR, net migration. Original source: population.un.org/wpp

Download
owid_historical.json(65 KB)

Historical population by country (1800-1949) from Our World in Data, sourced from Gapminder/Maddison Project. Estimates use modern borders. Original source: ourworldindata.org/population

Download

Model Code

population_real.py(31 KB)

Core simulation model — cohort-component population projection using real UN data. Includes Phase 1 (UN data 2024-2100) and Phase 2 (stochastic post-2100 extension). Requires PyTorch + CUDA.

Download
population_scenarios.py(25 KB)

Scenario projection engine — extends the base model with TFR override capability for running alternative fertility scenarios.

Download
population_multigpu.py(32 KB)

Multi-GPU orchestration — distributes Monte Carlo trajectories across 8 GPUs using ThreadPoolExecutor. Each GPU runs independent batches with local histogram accumulation.

Download
population_tfr_sweep.py(12 KB)

TFR sweep script — runs 100 TFR scenarios (0.1-10.0) with checkpoint/resume support. One scenario per GPU, 8 in parallel. Adaptive histogram bins for low- and high-population ranges.

Download
run_verified_scenarios.py(4.7 KB)

Script to reproduce original 6 named scenario results. Requires GPU with 24GB+ VRAM.

Download

How to Reproduce

# Requirements: Python 3.10+, PyTorch 2.x with CUDA

# Hardware: NVIDIA GPU with 24GB+ VRAM (single GPU) or 8× A100 (full sweep)

pip install torch pandas numpy

python population_real.py --validate # Should match UN within 1.6%

python population_multigpu.py --scenarios 10M # 6 scenarios on multi-GPU

python population_tfr_sweep.py # 100 TFR scenarios × 10M trajectories

License

All simulation code, results, and derived data are released under the MIT License. Source data from UN, OWID, Pew Research, and CIA Factbook retain their original licenses. UN WPP data is public domain. OWID data is CC-BY. Pew Research data requires attribution.