The Bank of Ghana (BoG) has stepped up its use of Artificial Intelligence (AI), machine learning and other advanced analytical tools to improve inflation forecasting, strengthen economic data gathering and support monetary policy decisions. First Deputy Governor Dr Zakari Mumuni said the technology is enabling the central bank to analyse increasingly large volumes of information and improve the accuracy and timeliness of its economic forecasts. He made the disclosure at the 4th Annual Statistics and Data Science Conference in Tamale, where he highlighted the growing importance of data science and advanced modelling in economic policymaking.
According to Dr Mumuni, AI-based tools are being used alongside conventional economic models to provide the Bank with a broader and more detailed understanding of developments that could affect the economy. He said the approach has contributed to improvements in the Bank’s inflation forecasts, including projections made before the release of official economic data. The Bank is also using machine-learning models to complement traditional econometric techniques in forecasting Gross Domestic Product (GDP) and conducting text-mining analysis. “We also employ machine-learning models to complement standard econometric models in forecasting GDP and performing text-mining analytics.” Dr Mumuni said the application of technology extends beyond forecasting to financial-sector supervision, where more granular and timely data is helping supervisors identify potential risks earlier.
He explained that the traditional approach, which relied heavily on static monthly spreadsheets and manual reconciliation, is gradually giving way to systems that allow information to be validated as it becomes available. This, he said, gives supervisors a more current picture of developments within financial institutions and strengthens the Bank’s ability to respond to emerging risks. The First Deputy Governor said the Bank also continues to rely on established economic tools, including econometric techniques and its Quarterly Projection Model within a Forecast and Policy Analysis System. These tools enable policymakers to identify emerging economic trends, assess risks and examine the potential consequences of different policy choices before decisions are taken. “Through econometric techniques and our Quarterly Projection Model within a Forecast and Policy Analysis System, we identify emerging trends, assess risks and consider the likely outcomes of different policy choices.”
Despite the growing role of technology, Dr Mumuni cautioned against treating AI and other analytical systems as substitutes for professional judgement. He said advanced technologies should enhance the quality of information available to policymakers rather than determine policy on their own. “Technology can strengthen our intelligence, but it does not remove the need for human judgment.” His comments reflect the Bank’s broader effort to adopt a more proactive approach to inflation management, following a directive from the Governor at Dr Mumuni’s swearing-in in February 2025. The Governor had called for the Bank to adopt a more proactive and precise approach to managing inflation through the use of advanced data analytics and artificial intelligence. Dr Mumuni said the rapid expansion of available data had changed the central challenge facing policymakers.
Rather than struggling primarily with a lack of information, policymakers now have to determine how to distinguish useful signals from the enormous volume of data available and convert them into reliable intelligence for decision-making. “The greatest challenge facing policymakers today is no longer a shortage of data, but rather turning an abundance of data into timely, reliable and actionable intelligence.” While emphasising the value of AI and big data, Dr Mumuni said the Bank has not abandoned traditional methods of collecting economic information directly from households and businesses. He said researchers continue to visit markets and communities across the country to monitor prices and gather information through business and consumer confidence surveys. Such fieldwork, he explained, provides policymakers with information about economic conditions across different parts of the country and helps prevent national policy decisions from being based solely on developments in the capital.
He noted that Research Department staff regularly spend time in communities, including Tamale, gathering first-hand information that feeds into the deliberations of the Monetary Policy Committee. The approach, he said, ensures that the Committee’s assessment of the economy reflects experiences from across Ghana. Dr Mumuni urged statisticians, researchers and policymakers to ensure that emerging sources of data strengthen rather than undermine established statistical practices. He cautioned that new datasets should not be allowed to replace properly designed, weighted and nationally representative measures. “New data should complement—not replace—properly weighted and nationally representative measures.” He also called for closer collaboration between researchers and policymakers, arguing that the relationship should be based on a willingness on both sides to challenge assumptions and scrutinise the evidence behind policy decisions.
He said researchers need to understand the practical questions confronting policymakers, while policymakers should remain receptive to research that challenges prevailing interpretations of economic data. The comments underscore the Bank of Ghana’s evolving approach to economic management, in which traditional statistical and econometric methods are increasingly being combined with AI, machine learning, big data and real-time information to strengthen the evidence available for monetary policy.
By: Joyce Owusu
