BY NADIA NTIAMOAH
Researchers at the Bank of Ghana (BoG) are advocating the adoption of big data and artificial intelligence (AI) to strengthen inflation monitoring, arguing that technology could help policymakers track price movements faster and make more timely decisions on the economy.
The proposed system, known as e-Inflation, uses automated data collection and artificial intelligence techniques to estimate headline, food and non-food inflation in near real time, potentially offering an alternative source of information alongside the country’s existing inflation measurement process.
The framework was developed by Bank of Ghana researchers Francis Mawuli Abude, Collins Kwesi Appiah and Julius B. Dasah, who examined the potential of digital technologies to improve the speed and efficiency of inflation measurement in Ghana.
In their working paper, Measuring Inflation in Ghana: Evidence from Big Data and Artificial Intelligence Techniques, published in October 2026, the researchers argue that the approach could provide policymakers with earlier indications of price pressures than the conventional system, which relies on periodic price collection and subsequent statistical analysis.
The findings could have implications for monetary policy decisions, particularly as the Bank of Ghana monitors inflation trends to support price stability.
AI System Tracks Prices Across 120 Websites
According to the researchers, e-Inflation gathers price information automatically from 120 websites and analyses approximately 500,000 price observations covering the same 307-item consumer basket used by the Ghana Statistical Service (GSS).
The system is designed to estimate changes in headline inflation, food inflation and non-food inflation by processing large volumes of online price data.
The researchers believe the approach could help address some of the time constraints associated with conventional inflation measurement by providing more frequent updates on price movements.
Under the existing process described in the paper, the GSS collects 47,877 price observations from 57 market centres to calculate inflation figures.
Official inflation data are typically released in the middle of the following month, meaning policymakers and businesses may have to wait before obtaining a comprehensive picture of price movements during the preceding month.
The proposed digital system, by contrast, could provide earlier estimates, allowing economic analysts to identify emerging price pressures before the official figures become available.
However, the researchers present e-Inflation as a potential complement to the existing system rather than a replacement for official inflation statistics.
Findings Show Strong Correlation With Official Inflation Figures
The researchers tested the framework against official inflation data covering the period from May 2021 to December 2023.
Their findings showed a correlation of 0.9941 between e-Inflation estimates and official headline inflation figures.
The correlation was 0.9836 for food inflation and 0.9944 for non-food inflation.
These results indicate that the estimates produced by the digital system closely tracked the movements recorded in the official data during the period studied.
The paper also reported average deviations of 0.3684 percentage points for headline inflation, 0.6909 percentage points for food inflation and 0.1375 percentage points for non-food inflation.
The researchers argue that the findings demonstrate the potential of big data and AI techniques to support inflation monitoring in Ghana.
Nevertheless, the strong correlations do not, by themselves, establish that the digital estimates are interchangeable with official statistics. Further validation would be needed to assess the system’s performance across different economic conditions and periods.
