Loss Models: From Data to Decisions, 4th Edition

CHAPTER 13

FREQUENTIST ESTIMATION

If a phenomenon is to be modeled using a parametric model, it is necessary to assign values to the parameters. The assignment could be done arbitrarily, but it would seem to be more reasonable to base the assignment on observations from that phenomenon. In particular, we assume that n independent observations have been collected. For some of the techniques it is further assumed that all the observations are from the same random variable. For others, that restriction is relaxed. There are two, essentially incompatible approaches to estimating parameters. This chapter covers the frequentist approach to estimation introduced in Section 10.2. An alternative estimation approach, known as Bayesian estimation, is covered in the next chapter.

The methods introduced in Section 13.1 are relatively easy to implement but tend to give poor results. Section 13.2 covers maximum likelihood estimation. This method is more difficult to use but has superior statistical properties and is considerably more flexible.