I am currently working with a multiple regression with three parameters; $\beta_1, \beta_2, \beta_3$. The individual T-test for each parameter contains large p values (>.05) while the p-value of the F-test for the null hypothesis that all coefficients are 0 is quiet small (approx. 0).
Is this suggestive of the issue suggested here of multicollinearity? And thus, would you reject or accept the null hypothesis, either for individual parameters or the model as a whole?