a425_form

# a425_form - Functional Form Qualitative Variables APS 425...

This preview shows pages 1–4. Sign up to view the full content.

Functional Form & Qualitative Variables APS 425 - Advanced Managerial Data Analysis (c) Prof. G. William Schwert, 2001-2010 1 APS 425 – Winter 2010 Functional Form and Qualitative Variables Instructor: G William Schwer Instructor: G. William Schwert 275-2470 [email protected] Topics • Transformations to linearity • Dummy variables Interaction variable • Interaction variables

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

View Full Document
Functional Form & Qualitative Variables APS 425 - Advanced Managerial Data Analysis (c) Prof. G. William Schwert, 2001-2010 2 Transformations to Linearity • Linear regression assumes that Y is linearly related to X 1 , X 2 , etc. with an additive error term • Scatter diagrams can help us understand whether the variables we are considering need to be transformed to create a linear relation – e.g., if we take logarithms of the variables (assuming that they are all positive numbers), this implies a constant elasticity relation between Y and X and log Y = 0 + 1 log X + => Y = 0 X exp( ) where log is the natural logarithm and exp(•) is the exponential function Transformations to Linearity Quantity Demanded at Different Prices \$6 \$8 \$10 \$12 rice per Unit Quantity Demanded at Different Prices (log transformations) 1.50 2.00 2.50 Price per Unit \$0 \$2 \$4 0 200 400 600 800 1000 1200 Units Demanded at a Given Price Pr 0.00 0.50 1.00 012345678 Log of Units Demanded at a Given Price Log of
Functional Form & Qualitative Variables APS 425 - Advanced Managerial Data Analysis (c) Prof. G. William Schwert, 2001-2010 3 Logs as Percent Changes • Suppose that log(Y) is linearly related to some variable X log Y = 0 + 1 X + Then the slope coefficient 1 measures the percent change in Y caused by a change in X Fact log (1 + r) r for |r|< 15 Fact: log (1 + r) r for | r | < .15 To the extent that the distribution of % changes is more likely to have a constant variance than the distribution of changes, the log transformation can help solve heteroskedasticity problems Logs as Percent Changes Logs as approximate percentage ln(1+r ln(1+r Excel function ln(•) is the r ln(1+r) r ln(1+r) -0.150 -0.163 0.150 0.140 -0.140 -0.151 0.140 0.131 -0.130 -0.139 0.130 0.122 -0.120 -0.128 0.120 0.113 -0.110 -0.117 0.110 0.104 -0.100 -0.105 0.100 0.095 -0.090 -0.094 0.090 0.086 -0.080 -0.083 0.080 0.077 Excel function ln () is the natural logarithm -0.070 -0.073 0.070 0.068 -0.060 -0.062 0.060 0.058 -0.050 -0.051 0.050 0.049 -0.040 -0.041 0.040

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.

## This note was uploaded on 03/03/2010 for the course APS 425 taught by Professor Schwert during the Spring '10 term at Rochester.

### Page1 / 13

a425_form - Functional Form Qualitative Variables APS 425...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online