我已經對此作出調查

我已經對此作出調查

結果 (繁體中文) 2:

If it was not this situation, QR was considered to be more robust than OLS. Table 2summarizes the multicollinearity test result of explanatory variables. It was clear that all VIF values ranging between 1.653 and8.553 were lower than 10, indicating that model(6) passed the multicollinearity test. As described in Table 3, all five test methods rejected the zero hypotheses at a significance level of 1%. This would imply that there was no unit root in the sequences of variables, that is, the sequences were stationary. Moreover, as it can be found from Fig. 3, the data points of almost all variables deviated from the blue fitting line .This would suggest that the variables were not normally distributed. Therefore, the QR model was more appropriate and reasonable than OLS estimation.

結果 (繁體中文) 3:

I've investigated this