Microsoft Word Thesis Gent (1). doc
Download 1.76 Mb. Pdf ko'rish
|
DPTX 2010 2 0 330292 0 110731
6.3 Empirical results
In order to interpret the results that macroeconomic shocks cause the credit risk, the impulse response function is used. The VAR model allows doing estimation of macroeconomic shocks and their impact on credit risk (NPL ratio). The VAR functions show the period of a variable in response (NPL ratio) to a shock in macroeconomic variables. This shows which, shock provokes the variation in an endogenous variable (NPL ratio). The execution is done by identifying the structural innovations from the 52 reduced form residuals, which in this case is used as Choleski-decomposition as it imposes zero restrictions. In addition, to verify that the lags of endogenous variable are significant to enhance the estimated performance of the other variables for the credit risk (NPL ratio), the Granger causality test was used. Specifically, these are pair wise tests of the null hypothesis that the coefficients of an endogenous variable in equation equal zero and thus Granger causes it. Hence, if the null hypothesis is rejected by a standard F-test, it means that there is inclusion of a specific variable in the equation of other variable The Granger causality tests on my model, shows that long term interest rate (IR_SA) does not Grangerly causes credit risk (NPL ratio), which means that long-term interest rate does not help to predict the credit risk (see table 2). Therefore, null hypothesis that long-term interest rate does not Grangerly causes NPL ratio is not rejected. The impact of inflation (L_CPI_SA) to credit risk (NPL ratio), is significant on 90% of confidence interval with the p-value 0.0647. So, null hypothesis has been rejected. Hence, inflation does Granger causes credit risk, and helps on predicting it. Important to note is that inflation has impact on credit risk, but credit risk does not Granger causes the inflation. In terms of exchange rate (L_ER_SA) as impact on credit risk Granger causality test shows that exchange rate does not Grangerly Causes credit risk. So the exchange rate does not help to predict the credit risk. Therefore, null hypothesis is not rejected. According to GDP (L_GDP) and unemployment rate (UNEMP), results show that both macroeconomic variables do not Granger causes the credit risk; even credit risk do not Granger causes shocks on GDP and unemployment rate. Thus it means that GDP and Unemployment rate, referring to Granger causality test do not help to predict the credit risk and null hypothesis are not rejected. In a nutshell, almost all macroeconomic variables are insignificant and do not Granger causes the credit risk, except the inflation, which has a significant impact on credit risk. Download 1.76 Mb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling