Last month's drop in domestic inflation provided the Bangko Sentral ng Pilipinas (BSP) some breathing…
BSP is looking into AI applications to improve its capabilities.
MANILA, Philippines – The Bangko Sentral ng Pilipinas (BSP) is investigating the use of machine learning (ML) techniques in natural language processing, nowcasting, and banking supervision.
“Over the years, central banks’ interest in machine learning has grown, owing to its potential to improve existing tools used for regular monitoring as well as its ability to uncover underlying relationships between data to better understand the economy and financial system,” said BSP Governor Benjamin Diokno.
Machine learning is a branch of artificial intelligence (AI) that involves algorithms that produce results based on data patterns.
Natural language processing is utilized at the BSP to turn text into data in order to provide quantitative summaries, such as the news sentiment index and the economic policy uncertainty index, which are currently being developed.
The BSP also uses machine learning to forecast regional inflation and domestic liquidity. These models are in addition to the BSP’s existing macroeconomic forecasting models.
The BSP wants to use machine learning techniques to improve its data validation processes and better identify anomalous data in banking supervision.
According to Diokno, machine learning has a wide range of applications in central banking, particularly when integrated with approaches from other fields like econometrics and network science.
Diokno also mentioned a few issues with machine learning methods. The black-box approach to ML is the most frequently stated restriction, which may make it difficult to interpret causal linkages in ML models.
ML algorithms, like traditional econometric methodologies, may have difficulty accurately anticipating tail risk or low likelihood occurrences.
The implementation of machine learning (ML) models will need investments in IT infrastructure and capacity creation, as well as a shift in organizational culture.
The BSP will continue to study ML applications that can be useful in the conduct of its important functions as the country builds a stronger and more technologically advanced Philippine economy, while carefully noting the inherent obstacles.