This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: •4/6/2011 •1 Lecture 20: April 6, 2011 • Material for today in Chapter 12: • Introduction to simple regression: bottom p.494middle p.504 • Next class: middle p. 504bottom p. 515 The following slide is a summary of last class. •4/6/2011 •2 Interpretation of r = .747 • A moderately strong • and • positive • linear association • between • household income • and • food expenditures Back To The Original Data: Let’s Do A Plot • X Y • 30 10 • 45 16 • 50 17 • 55 20 • 70 17 Back To The Original Data: Axes and Units • X Y • 30 10 • 45 16 • 50 17 • 55 20 • 70 17 •4/6/2011 •3 Back To The Original Data: Plot the 2 Means • X Y • 30 10 • 45 16 • 50 17 • 55 20 • 70 17 Back To The Original Data: Axis for Each Mean • X Y • 30 10 • 45 16 • 50 17 • 55 20 • 70 17 Back To The Original Data: Plot the Five Pairs • X Y • 30 10 • 45 16 • 50 17 • 55 20 • 70 17 •4/6/2011 •4 Back To The Original Data: Regions I  IV • X Y • 30 10 • 45 16 • 50 17 • 55 20 • 70 17 Limitations of Linear Correlation • It measures the degree to which variables are linearly associated . • It says nothing about – Causality – Functional relationships • Example: Crop yield (on the vertical axis) vs. fertilizer application rate (on the horizontal axis) – This relationship is parabolic, i.e., an upsidedown U. – The correlation coefficient would be close to zero. – To see this, make use of the regions in the previous slide and note what happens when you » calculate the products within each region, and then » sum over all of the products.» sum over all of the products....
View
Full
Document
This note was uploaded on 10/13/2011 for the course FINOPMGT 250 taught by Professor Kouzehkanani during the Spring '08 term at UMass (Amherst).
 Spring '08
 KOUZEHKANANI

Click to edit the document details