Consider the data forbes.data,
which gives ancient numbers for a sample of fortune 500 companies (at
the time). The response variable is Sales and of particular
interest is whether the sector variable has any effect.
- Do any of the variables need to be transformed to ensure a proper
analysis? If so, make the appropriate transformations.
- After this, compute the variance inflation factors for each of the
X variables.
- Build (an) appropriate model(s) for Sales. Pay attention to
transformations, variable selection, and outliers. Does sector have an
effect?