Where 1 /2 t −α is the (1 / 2) −α th percentile of the t-distribution with n+m-2 degrees of
freedom. Note that T1 has an exact t-distribution with n+m-2 degrees of freedom.
In the case where the two variances differ, i.e. 2 2 σ σ x y
≠ , the confidence interval for θ
is constructed using the pivotal quantity T2 ,