Lecture 21 2010

Lecture 21 2010 - % change in Y. 2 C. Non-Linear Models 1....

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
1 V. Extensions of Multiple Regression A. Dummy Variables B. Scaling Variables C. Non-Linear Models 1. Basic Idea for Estimation . 2. Models that don’t require logs 3. Models that require log transform . B. Scaling Variables 1. Scaling the Dependent and Independent Variables : The effects of scaling by a known constant are perfectly predictable: Suppose we multiply Y by ‘a’ and X by ‘b’ ( ) * 1 1 2 2 2 ( ) ˆ ˆ ( ) i i i i i i b x a Y x Y b a a b x b x b β = = = Scaling variables by constants has no real consequences. 2. Indexes – the case of an unknown scale. The independent variable, and possibly the dependent variable, may be “indexes.” Examples: Consumer Price Indexes. You have no idea about the “base year” value. Indexes – pure numbers, no “units.” Solution – calculate elasticities , pure numbers. A 1% change in X causes an
Background image of page 1

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

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: % change in Y. 2 C. Non-Linear Models 1. Basic idea : Make the model look linear in the parameters. Then estimate using OLS. 2. Models that dont require a log transform . Quadratic form: Examples: Interpretation: C. Non-Linear Models Inverse form: Y = + 1 (1/X) + u Example: Interpretation: C. Non-Linear Models 3. Models that require a log transform . Cobb-Douglas : log-log models. Transform using natural logarithm. Interpretation: 3 3. Models that require a log transform . Semi-Log models : log-linear Transform using natural logarithm. Parameters: 1 1 2 2 exp X X u Y + + + = 3. Models that require a log transform . Semi-Log models : linear-log. 1 2 1 2 exp exp Y u AX X = Transform using natural logarithm. Parameters:...
View Full Document

This note was uploaded on 12/08/2011 for the course ECON 312 taught by Professor Daniellass during the Winter '10 term at UMass (Amherst).

Page1 / 3

Lecture 21 2010 - % change in Y. 2 C. Non-Linear Models 1....

This preview shows document pages 1 - 3. Sign up to view the full document.

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