Financial extrapolation needs to be based on one or more focal variables, although not necessarily all the variables in an analysis. Most important in financial
extrapolation is to consider whether the focal variables can be valued with regard
to revenue or cost, as we discuss later.
When considering statistics that focus on associations between multiple variables the focal variable would often be a dependent variable, which is a construct that we use other variables to explain or predict.
In Chapter 8 we considered some simple association variables like correlation; in
later chapters we considered techniques like regression, which seek to explain or
predict certain variables.
There may also be other variables in the analysis, such as predictors of the focal
variable in a regression.
The first step involves using statistics to assess either the level of a focal variable
or the expected or actual change in the variable. What do I mean by this? There are, broadly, two extrapolation situations:
-
Estimated level of a variable: Descriptive statistics like the average or median, estimate the level of a focal
variable. In addition, more complex analyses can also lead to a final statistical
conclusion that is the level of a variable.
-
Estimated change in the focal variable: Some statistical analyses can also be used to predict expected change in a focal
variable. A regression slope is an example: it measures the expected change in the
dependent variable that is associated with a one-unit increase in the independent
variable.
To illustrate the difference between static and change situations, let’s start with
the simple example.