Chapter 2 Interval Estimation

In the previous chapter, we discussed the problem of point estimation, that is, finding something like best estimates for the values of unknown parameters. Simply quoting such numbers as the result of the measurement is not entirely satisfactory, however. At the very least, this lacks any idea of how precise the measurements may be. We now rectify this deficiency, with a discussion of interval estimation. The idea is to give a range of values, that is, an interval, for the parameter with a given probability content. The interpretation of the probability depends on whether frequentist or Bayesian approaches are being used.