Past exam of the mathematics course of the University of Cambridge / 2005 / ii / Paper 2 / 27I (Principles of Statistics)

Let $Q=P_1-P_0$ and $R=I-P_1$. For nested subspaces these are orthogonal projections onto ${\Theta }_1\cap {\Theta }_0^{\perp }$ and ${\Theta }_1^{\perp }$, of ranks $d_1-d_0$ and $d-d_1$. Under the null, both annihilate $\mu$. In an orthonormal basis adapted to ${\Theta }_0$, these two spaces and their complement, the coordinates of $X-\mu$ are independent $N(0,{\sigma }^2)$. Consequentlyand the two variables are independent. Equivalently the joint probability density of the two unscaled sums $a,b>0$ factors into gamma densities of respective shapes $(d-d_1)/2,(d_1-d_0)/2$ and common scale $2{\sigma }^2$.

The nuisance variance cancels inReject for a value above the upper $\alpha$ critical quantile to obtain a level-$\alpha$ test. Under a genuine alternative the numerator is noncentral chi-square with noncentrality $\Vert Q\mu {\Vert }^2/{\sigma }^2$, while the denominator retains its central law, explaining the upper-tail choice.

Assume $I,J\ge 2$. The additive two-way model has dimension $1+(I-1)+(J-1)=I+J-1$; the null model, with no exam effects, has dimension $I$. Write row, column and grand means as ${\overline{X}}_{i\cdot },{\overline{X}}_{\cdot j},{\overline{X}}_{\cdot \cdot }$. Orthogonal projection onto the full model gives fitted values ${\overline{X}}_{i\cdot }+{\overline{X}}_{\cdot j}-{\overline{X}}_{\cdot \cdot }$, while the null fit is ${\overline{X}}_{i\cdot }$. The added exam sum of squares and full residual sum of squares areThe preceding orthogonal-projection result gives independent null chi-squares with $J-1$ and $(I-1)(J-1)$ degrees of freedom. ThereforeReject the no-exam-effect hypothesis for a large value. The printed alternative repeats the identifying constraint ${\sum }_j{\beta }_j=0$; a genuine alternative additionally has at least one nonzero ${\beta }_j$. No interaction term is being fitted; the independent equal-variance Gaussian error assumption is what makes this exact test valid.