Study Questions, Class 12

1. How is the model for paired observations different from all the other sampling models we have been looking at?

The parent distribution is bivariate, not univariate.

2. You are generating a confidence interval for the difference of two means. You do not know the variances, and you cannot assume them to be equal. In one group, there are 12 samples. In the other, there are 9. What must or can you do? Assume the parent distributions are normal.

Randomly eliminate 3 of the 12 samples so that the n's are equal for the two groups.