Scribd
Upload
3 views

Chapter 6 - STATA

The document provides a hands-on guide to using Stata for estimating discrete choice models, including the Linear Probability Model, Logit Model, Probit Model, Multinomial Logit/Probit Model, and Ordered Logit/Probit Model. It outlines specific commands for each model and instructions for computing marginal effects using various datasets. The document emphasizes the comparison of results between different modeling approaches.

Uploaded by

abenezer
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views

Chapter 6 - STATA

The document provides a hands-on guide to using Stata for estimating discrete choice models, including the Linear Probability Model, Logit Model, Probit Model, Multinomial Logit/Probit Model, and Ordered Logit/Probit Model. It outlines specific commands for each model and instructions for computing marginal effects using various datasets. The document emphasizes the comparison of results between different modeling approaches.

Uploaded by

abenezer
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

5/20/2023

Hands-on Using Stata


Sisay Deebebe (PhD)

Addis Ababa University

Discrete Choice Model


The Linear Probability Model

• Suppose we want to estimate

where yi = probability of owning a car, xi = income

• Use the data under file name: car owner


• In the data, car & income stand for Y1 & X2, respectively.
• Estimate using the command:
• reg car income
• Interpret the results.

1
5/20/2023

Discrete Choice Model


The Logit Model

• Estimate the model using logit:


 logit car income

• To comupte the marginal effects, estimate first the scale effect:


mfx compute

Discrete Choice Model


The Probit Model
• Estimate the model using probit:
 probit car income
• Compute marginal effects:
 mfx compute
• Compare logit & probit

2
5/20/2023

Discrete Choice Model

• Consider the following model where one of the covariates is


binary

WhereYi = Grade (=1 if grade in later course intermediate


economics- is higher than the grade in previous courses-principle
of economics)
X1 = GPA
X2 = Pretest score
X3= a dummy for participation in a new teaching method
(PSI- Personalized System of Instructions)

 Use the data under the file name: Binary Choice


 Variables grade, gpa, tuce, and psi represent Yi , X1 , X2 , X3
,respectively.
5

Discrete Choice Model

 Estimate the model using OLS (LPM)


 reg grade gpa tuce psi
 Estimate the model using Logit:
 logit grade gpa tuce psi
 Compute the marginal effects:
 mfx

3
5/20/2023

Discrete Choice Model


The Multinomial Logit|Probit Model
• Use the data under file name: multinomial
• Estimate the Multinomal model where the dependent
variable is status.
 mlogit status educ exper expersq black if year==87
 Mlogit status educ exper expersq black, base(1), if year==87
 Compute the marginal effects
 mfx, predict(p outcome(1))
 mfx, predict(p outcome(2))
 mfx, predict(p outcome(3))

Discrete Choice Model


The Multinomial Logit|Probit Model…

• Estimate the model using multinomial probit:


 mprobit status educ exper expersq black if year==87

• Compute the marginal effects


 mfx, predict(p outcome(1))
 mfx, predict(p outcome(2))
 mfx, predict(p outcome(3))

4
5/20/2023

Discrete Choice Model


The Ordered Logit|Probit Model

• Assume that the dependent variable we used in the


multinomial model is ordered (3>2>1).
• Estimate the model using ordered logit:
• ologit status educ exper expersq black if year==87
• Compute the marginal effects
• mfx, predict(p outcome(1))
• mfx, predict(p outcome(2))
• mfx, predict(p outcome(3))

Discrete Choice Model


The Ordered Logit|Probit Model…

• Estimate the model using Ordered Probit:


• oprobit status educ exper expersq black if year==87
• Compute the marginal effects
mfx, predict(p outcome(1))
mfx, predict(p outcome(2))
mfx, predict(p outcome(3))

10

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy