Logo Lanfrica

mosesx3m/TS_Academy_Capstone_Project

Domain:

agriculturesocioeconomic

Record type:

project
Creator:
mos
Host:
A Time Series Analysis & Forecasting System For Household Food Security In Nigeria. # TS_Academy_Capstone_Project ## Nigerian Food Price & Food Inflation Forecasting ### TS Academy — Track 3: Time Series Analysis | Capstone Project --- ## Project Overview Nigeria has experienced severe food price inflation over the past decade, with conditions worsening significantly following the removal of the fuel subsidy in May 2023. This capstone project builds a **time series forecasting system** for two critical economic indicators: - **Basket Price** — a weighted composite of 14 staple food commodities reflecting the real cost of a typical Nigerian household's monthly food spend - **CPI Food** — the official Consumer Price Index for food, as published by the National Bureau of Statistics (NBS) By modelling both indicators simultaneously and linking them through a **cascade forecasting architecture**, we produce 6-month ahead forecasts that can inform household planning, policy decisions, and economic analysis. --- ## Problem Statement Food affordability is one of the most pressing economic challenges facing Nigerian households today. Between 2022 and 2024, food prices surged by over 80% in nominal terms, driven by: - The removal of the petrol subsidy in June 2023, which caused immediate and severe cost-of-living shocks - Persistent naira depreciation against the US dollar, raising the cost of imported food and inputs - Supply chain disruptions and structural inefficiencies in local food markets Despite the severity of the crisis, reliable short-term forecasts for food prices remain scarce. Existing tools either rely on lagged official data or fail to account for structural breaks caused by policy changes. This project addresses that gap by developing **data-driven time series models** that: 1. Explicitly model the 2023 structural break using shock dummy variables 2. Incorporate macroeconomic drivers (fuel price, exchange rate) as exogenous regressors 3. Validate a linked architecture where basket price forecasts improve CPI food predictions 4. Q …