Economic Crisis Treatment Based on Artificial Intelligence

Abstract

Abstract: There are many possible causes of an economic crisis—a financial downturn, a banking meltdown, political strife (e.g., the Russia-Ukraine war), or a health-related catastrophe (e.g., Covid-19). Some of these crises are expected, while others are “bolts from the sky.” However, what is certain is that all these crises, whatever their cause, have a negative impact on global gross domestic product (GDP). If we can identify the components of output that have the most impact in an economic crisis, we might be able to mitigate its effects. Therefore, this paper uses machine learning algorithms to determine how the components of expenditure and sectoral value-added approach impact global GDP. The gradient boosting algorithm is the most accurate model for predicting and determining the impact of independent variables on a dependent variable. The results indicate that government spending has the largest effect on global GDP, accounting for 68.3% of the impact. The economic sector with the most impact on global GDP is the service sector, which affects global output by 42.3%, followed by the agricultural sector at 30.2%. Thus, stimulating government spending and the service sector may reduce the negative effects of an economic crisis.