EiA2030-ex-ante/profitability-gaia

Lime profitability analysis for the GAIA trials (Ethiopia, Rwanda, Tanzania): yield response functions, NPV, sensitivity and risk analysis, plus an interactive Shiny explorer # Lime Profitability Analysis — GAIA Trials (Ethiopia, Rwanda, Tanzania) An end-to-end, reproducible framework for estimating **crop yield response to an agronomic input** (here: agricultural lime on acid soils) and translating it into **farm-level economics** — first-year profit, multi-year net present value (NPV), break-even prices, probability of profit under price risk, and distributional (quantile) returns. The repository contains two standalone components: | Component | What it is | Start here | |---|---|---| | `analysis/` | A numbered R pipeline (`00`–`09`) that goes from raw trial data to publication-ready tables and figures | `analysis/README.md` | | `shiny_app/` | An interactive Shiny explorer for the same data: response curves, NPV, Monte Carlo, risk, and adoption scenarios | `shiny_app/README.md` | Each component runs from its own folder with its own data — you can copy either one out of the repository and it will still work. Although built for the GAIA lime trials, the pipeline is written to be **reused as a template for other input-response profitability studies** (fertilizer, improved seed, irrigation) — see Adapting this framework. --- ## 1. The study in one paragraph Soil acidity constrains crop production across the East African highlands. The GAIA trials applied four lime rates — 0 (control), 1, 2.5, and 7 t/ha — on farmer fields in Ethiopia, Rwanda, and Tanzania, measuring yields for maize, beans, wheat, soybean, and faba bean (~1,700 year-1 observations across 13 sites). This analysis fits a **crop-level yield response function** to lime (a linear mixed model pooling all sites, with soil and rainfall covariates absorbing site heterogeneity), then asks the economic questions that matter for farmers and policy: *At what lime rate is profit maximized? How long until liming pays back? How robust is profitability to crop and lime prices? Which fields should be targeted first?* ## 2. Repository structure ``` ├── analysis/ …