Stat 112: Wrapup of simple regression

• Administrivia:
  • JMP Homework due today.
  • I listed on the schedule three more data sets to analyse from the book.
  • These are for "fun" not to be turned in. You probably should work in your group of two people to do these. Please do them this weekend if possible.

• Cottages example (jmp data)
  • Assumptions: notice influential outlier. Should we delete it or not? Both have their advantages.
  • Consider building a 3000 square foot cottage
  • Would it be profitable?
  • Make prediction

• Housing prices (jmp data)
  • Influential point (outlier and leveraged)
  • Try it with and without
  • Even without the residuals look curved

  • so try a log plot (give prediction intervals)
  • interpret the log plot
  • generate an actual log plot (give prediction intervals)

  • How about loging the X axis also? (called log-log plot)
  • Generates an elasticity
  • slope is in %change Y/%change X

• Common problems in regressions (handout)

Last modified: Thu Feb 10 11:25:09 2000