e-Inflation — the Bank of Ghana's In-House AI and Big-Data Nowcasting System for Monetary Policy
Ghana
The Bank of Ghana built an in-house 'e-Inflation' nowcasting methodology and paired it with machine-learning models to sharpen GDP and …
Kenya · Nairobi · See the Kenya profile · See the Nairobi profile
Top 35% 67/100 · Ask Evidence Copilot about this practice
An IMF-assisted nowcasting model gives Kenya's central bank weekly-updated GDP growth estimates from mobile-money, electricity, trade and remittance data, cutting the usual three-month lag and beating prior CBK and IMF forecasts out-of-sample.
Kenya's official GDP figures, like most developing economies', are published more than three months after each quarter ends — too slow to guide interest-rate decisions in real time. Starting with an IMF Monetary and Capital Markets Department technical-assistance mission to Nairobi in April 2023, the Central Bank of Kenya (CBK) built a nowcasting framework to close that gap, formally documented in IMF Working Paper 2026/032, "Nowcasting GDP Growth for Kenya."
The model continuously updates a near-term GDP growth estimate as new data arrive: mobile-money transaction volumes (available within about three weeks), then remittances, then trade, money-supply, tourism and electricity data (available within roughly 40 days) — around half the time of the official release. Key swings in the nowcast since 2024 have come from electricity and vehicle-production data, digital-money flows, and goods trade.
The IMF working paper reports that the nowcasting framework "performs well out-of-sample compared to alternative forecasts by the CBK and the IMF" — an explicit, published comparative accuracy result rather than an unverified claim. VoxDev and Business Daily Africa independently covered the same IMF technical work for a general-readership and a business audience respectively. The 2024 IMF Technical Assistance Report that preceded the paper also flagged data-quality caveats: Kenya's underlying business surveys needed strengthening for the nowcast to stay reliable, work the CBK and IMF describe as still ongoing.
Read the full analysis: https://voxdev.org/topic/macroeconomics-growth/tracking-kenyas-economy-real-time
Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.
Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.
Where this practice's information was retrieved from, and when.
Ghana
The Bank of Ghana built an in-house 'e-Inflation' nowcasting methodology and paired it with machine-learning models to sharpen GDP and …
Germany
Since late 2022 the ECB has run a quantile-regression-forest machine-learning model in its monetary-policy toolkit. In Q2 and Q4 2025 …
Australia
The RBA built an NLP tool mining 25 years of ~22,000 business liaison notes; added to wage-growth nowcasting models, it …
India
Reserve Bank of India economists built a hybrid ML and time-series model — adding financial-market data and an uncertainty index …
Open full copilot Grounded in cited practices — always check the sources.