Plain-language summary
- What it solves
- Fertiliser prices depend on natural gas, but the path between the two is invisible.
- Who it can serve
- Anyone following agricultural or fertiliser markets, and anyone studying cost propagation along a chain.
- What it produces
- The theoretical cost of ammonia and urea, the gap to market price, and a one-month forecast checked afterwards.
Link feedstock, chemical balance and falsifiable forecasting while keeping theoretical cost, market benchmark and forecast separate.
3 · What it produces
Implied NH₃/urea cost, model-market spread, interval forecast and matured track record.
The cards expose multiple views of the same workflow so result, method, evidence and history remain distinct.
4 · How it works
Physical cost chain + econometric transmission + maturation
1TTF, World Bank Pink Sheet, ECB FX, lags/seasonality; explicit HHV/LHV
2deterministic NH₃/urea chain + Ridge/Gradient Boosting M+1 forecast
3walk-forward, persistence, seasonality, gas-only and physical baselines; MAE/RMSE/bias/coverage/skill
4Persistence, rendering and governed publishing.
5 · Data and signals
| Element | Detail |
|---|---|
| Data and signals | TTF, World Bank Pink Sheet, ECB FX, lags/seasonality; explicit HHV/LHV |
| Industrial transferability | fertilizers, feedstocks, cost pressure and procurement intelligence |
| Research | lags, elasticity and ML-vs-physical-model comparison |
| Interpretation condition | global benchmark kept distinct from modelled European cost |
deterministic NH₃/urea chain + Ridge/Gradient Boosting M+1 forecast
walk-forward, persistence, seasonality, gas-only and physical baselines; MAE/RMSE/bias/coverage/skill
Python, pandas, scikit-learn, SQLite, Matplotlib
mistaking a global benchmark for European cost or mixing nowcast and forecast
Workflow technical dossier
Methods, checks, stack and transferability conditions.
Data and signals
TTF, World Bank Pink Sheet, ECB FX, lags/seasonality; explicit HHV/LHV
Inference / calculation
deterministic NH₃/urea chain + Ridge/Gradient Boosting M+1 forecast
Validation
walk-forward, persistence, seasonality, gas-only and physical baselines; MAE/RMSE/bias/coverage/skill
Stack
Python, pandas, scikit-learn, SQLite, Matplotlib
Limitations
mistaking a global benchmark for European cost or mixing nowcast and forecast
See recent outputs
The Publications page collects local previews and, on request, the public feed.
Potential transferability
Have a problem similar to Ammonia & Fertilizer Chain?
A first discussion can start from the objective, available data and success criterion without sending confidential information. A result observed here is not assumed to transfer automatically to another context.