4-13-09 - Make sure your theoretical model and empirical...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Economic Statistics – Class Notes – 4/13/09 Recall – Multicollinearity – Two or more independent variables are linearly related – Estimators: 1.) Should be unbiased 2.) Should be efficient Multicollinearity does not affect bias or efficiency of our estimators Major Effect:
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
First Correction : Leave out one of the affected variables – if theory permits Example: Unemployment rises during a recession. o So instead of including both variables (recession and unemployment), just use unemployment to improve the quality (significance) of your model.
Background image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Make sure your theoretical model and empirical models are consistent Second Correction : Find a proxy (a substitute) for the affected variable another variable that measures the same thing as the original variables, but is less linearly related to the other Example : Affected variable = interest rate . Replace the prime interest rate with the home mortgage interest rate. SEE THE IN-CLASS WORKSHEET ON MULTIPLE REGRESSION...
View Full Document

Page1 / 2

4-13-09 - Make sure your theoretical model and empirical...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online