Question

If a quadratic regression model is Yi =β0 + β1 X1i +β2 X1i2 + ‎ε, what should be the alternate hypothesis (H1) for testing the quadratic effect?

Answer 1

The alternative hypothesis (H1) for testing the quadratic effect would be that the coefficient β2 is not equal to zero. In other words, the quadratic term X1i2 has a significant effect on the dependent variable Yi.