16dec Flashcards
R²=0.7930 (goodness of fit)
The coefficient of determination is not very high (R²=0.7930), but it is also not very low. It indicates that 79.30% of the sample variability in Y is explained by its linear relationship with the sample variability in X2 and X3.
p-value < alpha =>
the model as a whole/the estimated coefficients is/are significant => continue working with the model.
𝛽̂₂ = 1.56×10⁻⁵ units
If X2 increases by 1 unit, while the X3 remains constant, it is estimated that the average Y increases by 1.56×10⁻⁵ units.
Contrast whether the linear specification that we have proposed is correct, indicating null and alternative hypotheses and conclusions from the result obtained.
Ramsey’s RESET test;
p-value < alpha => reject H0 => linear specification is incorrect.
b) What consequences could have made a specification error in the functional form have on our model?