The effect of Economic Growth on Non-performing Loans in Iran Based on Singular Spectrum Analysis and Nonparametric Regression Approaches

The amount of non-performing loans is one the indicators for assesssing banks credit risk and its high values is a sign of unhealthy of banking system. The aim of this study is to evaluate the impact of economic growth on NPLs by applying new nonparametric and robust approaches in time series analysis. For this purpose, an econometric model is designed on factors affecting NPL which includes three variables related to economic growth and a variable which is domestic credits. Quarterly data were used between the first quarter of 2009 to the first quarter of 2018 which have been gathered from the website of the Iran’s Ministry of Economy Affairs and Finance. Based on nonparametric approaches considered for analysing data, sub-space-based unit root test was performed to evaluate the stability of series and simple non-parametric regression model was performed for modelling propuses. In this paper, the relationships between variables were estimated using second-order Gaussian kernel in multivariate non-parametric regression. According to the results of the empirical analysis in Iran, there is a causal relationship between the non-performing loans and the total amount of loans of Iranian private banking sector. The SSA causality test shows that this relationship is evident. Gross Domestic Product (GDP) at fixed prices, Public Sector Expenditure (PS) and Private Sector Expenditure (PSE), Domestic Credit Volume (CV) are the most important subdivisions of economic growth. According to the results, public sector expenditure has an opposite effect and the increase in credit volume has a direct effect on increasing NPL.