Parallel Computing for Data Science

3.9. EXAMPLE: PARALLEL DISTANCE COMPUTATION 65

Many clustering algorithms make use of distances, for example. These tend

to be complex, so in order to have a more direct idea of why distances are

important in many statistical applications, consider nonparametric regres-

sion.

Suppose we are predicting one variable from two others. For simplicity

of illustration, let’s use an example with concrete variables. Suppose we

are predicting human weight from height and age. In essence, this involves

expressing mean weight as a function of height and age, and then estimating

the relationship from sample data in which all three variables are known,

often called ...