ABSTRACT
In this chapter we build on our previous discussion of sampling from a population
to obtain information about that population. We discuss in more detail the concept
of a statistical model for a random sample, which is a probability distribution that
depends on characteristics of the population and the way in which we took our sam-
ple from the population. The focus of the chapter is estimation of unknown quantities
associated with this probability distribution, called parameters, which represent char-
acteristics of interest within the population. We limit our discussion in this chapter
to properties of given estimation procedures and criteria for assessing their useful-
ness. In subsequent chapters we will broaden this to discuss general methods for
deriving estimation procedures, and for drawing inferences about the unknown pa-
rameters. The concepts are briefly illustrated using our disease prevalence example
from Chapter 1, and are then exemplified in greater detail using an extended example
on modelling a continuous risk factor.
