Result and Discussion

Multi-Colleniority Tests

To test the existence of multi-colleniority phenomena between model variables, Pearson correlation coefficients calculated between independent (predictor) variables, the results of testing multi-colleniority between independents variables were explained by correlation matrices and VIF test as follows.

Table 4 shows that the maximum value of correlation coefficient was between (loss aversion) and (Risk Perception), otherwise the values were less than or equals (0.355), which means there were no perfect relationship between variables. In the statistical literature the value (0.80) and more considered as an indicator of multi-colleniority existence.

Table 4: Multi-Colleniority Test For Predictor Variables
Variable Loss Aversion Overconfidence Herding Risk Perception
Loss Aversion 1.000
Overconfidence 0.144 1.000
Herding 0.169* 0.147 1.000
Risk Perception 0.355** 0.106 0.312** 1.000

Note: (**) Significant at 0.01; (*) Significant at 0.05.

To ensure the above result, the Variance Inflation Factor (VIF) was calculated, the results are illustrated in the following Table 5.

Table 5: Vif For Independent Variables
Variable VIF Tolerance
Loss Aversion 1.163 0.860
Overconfidence 1.038 0.963
Herding 1.128 0.887
Risk Perception 1.238 0.808