A MDS and PLS ’s Combined Approach for Exploring, Visualizing and Explaining Tunisian Banks’ Systemic Risk

Abstract:

In this paper, we provide a systemic risk cartography for Tunisian banks using Multi Dimensional Scaling (MDS) technique. To do so, we construct a presentation 1 and Presentation2 ’ based matrix that reflects systemic interdependency structure between banks. This latter depicts the contribution and the exposure of banks to systemic risk as well as the extent of their idiosyncratic risk. The recovered map reveals that public banks are taking the lead of the most implicated banks into systemic risk, followed closely by two of the most important private banks BIAT and UBCI. This same technique allowed putting forward a Systemic Risk Implication Composite Index (SRICI). Using this index we undertook a Partial Latent Structure regression (PLS) that implicates multiple sources of systemic risk in order to analyze the key drivers of the implication of banks in systemic risk. We show that systemic implication of Tunisian banks is highly dependent on the size of the financial institution, its liquidity, its direct exposure from interbank lending and its technical efficiency.

nsdlogo2016