Multivariate Monitoring of Gross Domestic Product and Inflation Rate in Ghana

Abstract

Multivariate control charts are statistical tools increasingly used for the simultaneous monitoring of multiple interrelated variables. This study applied Hotelling T², multivariate cumulative sum (MCUSUM), and multivariate exponentially weighted moving average (MEWMA) control charts to jointly monitor Gross Domestic Product (GDP) and inflation rate in Ghana, aiming to detect both small and large shifts in the mean vector of these variables. Annual data for the period 1973–2022 were obtained from the Bank of Ghana. Results indicate that the Hotelling T² chart flagged out-of-control points in 1976, 1980, 1982, 2012, and 2013, primarily reflecting moderate-to-large shifts in GDP and inflation. The MCUSUM chart detected a deviation in 1982, while the MEWMA chart identified out-of-control points in 1980, 1982, and 2013, capturing subtle but persistent changes. Comparative analysis suggests that Hotelling T² is most effective for detecting moderate-to-large shifts, whereas MCUSUM and MEWMA provide complementary sensitivity to smaller or time-weighted changes. This study is novel in applying multivariate SPC techniques to Ghana’s macroeconomic indicators, offering a proactive framework for monitoring GDP and inflation jointly. Integrating such charts into the Bank of Ghana’s economic monitoring tools could facilitate earlier detection of macroeconomic deviations and support more informed policy responses.