I'm studying for a CFA exams right now.  One of the study sessions is
about statistics.
The term "ordinary least squares" is discussed in the context of using ordinary
least squares to estimate a regression equation.  What are ordinary  least squares
and how are they used to estimate regressions?

Hey there:

This, again, is a question that the CFA cannot ask you.  


The term "ordinary least squares" is just a pretentious way of saying
"regression."  Regression, in the most general "plain vanilla" sense
is derived by minimizing the "sum of squared errors" of a fitted line.
 This minimization is essentially minimizing "variance."  It is not
something you will ever be asked to do on a CFA exam.