A dependent variable is regressed on two independent variables. It is possible that the hypotheses H0: 1

Question:

A dependent variable is regressed on two independent variables. It is possible that the hypotheses H0: β1 = 0 and H0: β2 = 0 cannot be rejected at low significance levels, yet the hypothesis H0: β1 = β2 = 0 can be rejected at a very low significance level. In what circumstances might this result arise?