The estimation problems discussed thus far were solved in the same general way. We first obtain
a statistic computed from the data that is an estimator of the population parameter of interest.
Then apply appropriate statistical theory to determine the sampling distribution of the statistic.
The sampling distribution can be described as the result of a *thought experiment* in which
we imagine obtaining every possible sample of size *n* from the population. We compute this
for each possible sample and then ask what is and appropriate model for the histogram of all those
values. That model should be a function of the population parameter. If the model includes
additional parameters, then those must be estimated from the data as well.

2016-02-04