4-17-09

# 4-17-09 - ” mean again 2 Detection Graphical Techniques...

This preview shows pages 1–7. Sign up to view the full content.

Economic Statistics – Class Notes – 4/17/09 c.) Autocorrelation – A situation in which error terms are linearly related to one another * There is a relationship between the error terms; they are supposed to be totally random If there is a relationship between the error terms, there might be way to find the relationship between the dependent variables Autocorrelation – usually associated with time series data. Time series data – data collected over time Cross-sectional data problem with cross-sectional data heteroscedasticity 1.) Effect OLS (ordinary least squares) estimators of are still unbiased, but NO LONGER EFFICIENT. o What does “ordinary

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

View Full Document

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

View Full Document

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

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ” mean again???? 2.) Detection Graphical Techniques Two ways to detect: a. Positive autocorrelation – We plot the error terms for periods “t” and “t – 1” Two ways to look at positive autocorrelation: NO AUTOCORRELATION – A RANDOM PATTERN b. The Durbin Watson (hypothesis) Test for 1 st Order Autocorrelation • With first order autocorrelation, error terms in period “t” are related to error terms in period “t – 1.” Epsilon(t) used to be the error, but NOW it is related to Epsilon(t-1), so we need another variable to represent the error u(t) For Monday: Look at MC (multicollinearity) autocorrelation...
View Full Document

## This note was uploaded on 06/03/2009 for the course ECON 203 taught by Professor Casler during the Spring '09 term at Allegheny.

### Page1 / 7

4-17-09 - ” mean again 2 Detection Graphical Techniques...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online