Show that the square of the expression for z in Exercise 12 equals so that the two tests are actually equivalent when the alternative hypothesis is θ1 
θ2. Note that the test described in Exercise 12, but not the one based on the χ2 statistic, can be used when the alternative hypothesis is θ1 2 or θ1 > θ2.
Exercise 12
Given large random samples from two binomial populations, show that the null hypothesis θ1 = θ2 can be tested on the basis of the statistic
