UNIVERSITY COLLEGE LONDON
DEPARTMENT OF ECONOMICS
Economics BSc (Econ)
Third Year
ADVANCED ECONOMICS
STATA USER GUIDE
Undergraduate
Rodrigo Antón García
London, 2022
, Rodrigo Antón García
STATA SUMMARY
Index
o 1 – Introduction. - Multiple Endogeneous Variables.
o 2 – Basics. • 6.5 – Limited Dependent Variable
• 2.1 – Stata Files Extensions. Models.
• 2.2 – Basic Commands. - Probit Regression.
• 2.3 – Syntax of Stata. - Logit Regression.
• 2.4 – Log Files. - Marginal Effects (PEA and APE).
• 2.5 – Controlling Output. - Panel Data.
• 2.6 – Saving Files. • 6.6 – First-Differencing Estimates.
o 3 – Exploring the Data. - Clustered Standard Errors.
o 4 – Creating New Variables. • 6.7 – Fixed Effects Estimates.
• 4.1 – Creating Dummy Variables. • 6.8 – Random Effects Estimator.
- Time Dummy Variables. o 7 – Other Estimates.
- DiD Estimator. • 7.1 – Difference in Means.
• 4.2 – Missing Observations. o 8 – Graphs.
• 4.3 – String Variables. • 8.1 – Histograms.
• 4.4 – Generation of Variables • 8.2 – Scatter Graphs.
(Extended). • 8.3 – Line Graphs.
• 4.5 – Drop (a) Variable(s). • 8.4 – Combining Graphs.
• 4.6 – Using Displayes Variables. o 9 – Distributions and Testing.
o 5 – Useful commands. • 9.1 – Student t and F Distributions.
• 5.1 – Prefix ‘by’. - Student t Distribution.
o 6 – Regressions. - F Distribution.
- Cross-Sectional Data. • 9.2 – Hypothesis Testing.
• 6.1 – OLS Regression. - t Statistic.
- Residuals. - F Statistic.
- Heteroskedasticity-Robust Errors. - Chow Statistic.
- Outliers. - Breusch-Pagan F Statistic.
- Marginal Effects. - Hausman Statistic.
• 6.2 – GLS (WLS) Regression. - Cragg-Donald Wald (First Stage) F-
Statistic (Weak Identification Test).
• 6.3 – Quantile Regression.
- Sargan–Hansen test or Sargan's J test
- LAD Regression.
(Overidentifying Restrictions Test).
• 6.4 – IV Regession
• 9.3 – Other Distributions.
- Multiple Exogeneous Variables.
- Binomial Distribution.
- Calculating IV from First Principles:
- Hypergeometric Distribution.
First-Stage and Reduced Form.
- Standard Normal Distribution.
- Multiple Instruments: Two-Stages
- Chi-Squared Distribution.
Least Squares (2SLS) Estimator.
1
, Rodrigo Antón García
o 1 – Introduction. o 2 – Basics.
Stata is a statistical package that includes a • 2.1 – Stata Files Extensions.
wide variety of capabilities, like data
management, statistical and econometric The following extensions are used/created by
analysis, graphics, etc. The user’s interface Stata:
includes the following:
- .dta: datasets that have been created
- Command Window: here we type all by Stata (using the command save).
the instructions (commands) that we ask - .do: programs (sets of instructions)
Stata to execute. Alternatively, a set of lines created with the Stata Do-File Editor.
of commands containing instructions can be - .dct: dictionaries; these are auxiliary
executed from the Do-Editor Window (See in files that are used to import data from text
the Stata Menu/Window/Do-file editor/New files.
do-file editor or just click on the icon - .smcl: log file (explained below) that
displaying a white paper and a pencil). Files can be read by Stata only.
created from the Do-editor (these are called - .log: log file that can be read by any
Stata programs) have the extension .do at the text editor.
end of the filename.
- Results Window: this space is used to • 2.2 – Basic Commands.
display all the results and output originated
by the commands that we have typed. To clear all the variables that Stata has
- Variables Window: shows all the stored in the memory (this is necessary for
variables currently stored in Stata’s memory. importing new data) we can type in the
We can also visualize these variables as in a Command Window:
spreadsheet by typing in the Command
Window browse or br followed by the clear
variables to be displayed (default shows all).
This is the Data Editor window. If we want When we need to learn the use of a command,
to make changes in these data, we will type what options it allows, or to see some
edit in the Command Window. examples of its use, we can type
- Command History: keeps a record of
all the commands used in each session. help name_of_the_command
- Current Working Directory: this
shows the current directory in the file system If we are unsure about the name of a
of your computer from where Stata will read command we are looking for, we can use
or save any files. Stata automatically seeks in search instead.
that directory programs, files, or anything
else it might request. This can be changed by search name_of_the_command
writing cd path_to_the_new_directory or
from the Stata menu: File/Change Working Any command in Stata that is preceded by a
Directory. star character ( ∗ ) will be regarded as
2
DEPARTMENT OF ECONOMICS
Economics BSc (Econ)
Third Year
ADVANCED ECONOMICS
STATA USER GUIDE
Undergraduate
Rodrigo Antón García
London, 2022
, Rodrigo Antón García
STATA SUMMARY
Index
o 1 – Introduction. - Multiple Endogeneous Variables.
o 2 – Basics. • 6.5 – Limited Dependent Variable
• 2.1 – Stata Files Extensions. Models.
• 2.2 – Basic Commands. - Probit Regression.
• 2.3 – Syntax of Stata. - Logit Regression.
• 2.4 – Log Files. - Marginal Effects (PEA and APE).
• 2.5 – Controlling Output. - Panel Data.
• 2.6 – Saving Files. • 6.6 – First-Differencing Estimates.
o 3 – Exploring the Data. - Clustered Standard Errors.
o 4 – Creating New Variables. • 6.7 – Fixed Effects Estimates.
• 4.1 – Creating Dummy Variables. • 6.8 – Random Effects Estimator.
- Time Dummy Variables. o 7 – Other Estimates.
- DiD Estimator. • 7.1 – Difference in Means.
• 4.2 – Missing Observations. o 8 – Graphs.
• 4.3 – String Variables. • 8.1 – Histograms.
• 4.4 – Generation of Variables • 8.2 – Scatter Graphs.
(Extended). • 8.3 – Line Graphs.
• 4.5 – Drop (a) Variable(s). • 8.4 – Combining Graphs.
• 4.6 – Using Displayes Variables. o 9 – Distributions and Testing.
o 5 – Useful commands. • 9.1 – Student t and F Distributions.
• 5.1 – Prefix ‘by’. - Student t Distribution.
o 6 – Regressions. - F Distribution.
- Cross-Sectional Data. • 9.2 – Hypothesis Testing.
• 6.1 – OLS Regression. - t Statistic.
- Residuals. - F Statistic.
- Heteroskedasticity-Robust Errors. - Chow Statistic.
- Outliers. - Breusch-Pagan F Statistic.
- Marginal Effects. - Hausman Statistic.
• 6.2 – GLS (WLS) Regression. - Cragg-Donald Wald (First Stage) F-
Statistic (Weak Identification Test).
• 6.3 – Quantile Regression.
- Sargan–Hansen test or Sargan's J test
- LAD Regression.
(Overidentifying Restrictions Test).
• 6.4 – IV Regession
• 9.3 – Other Distributions.
- Multiple Exogeneous Variables.
- Binomial Distribution.
- Calculating IV from First Principles:
- Hypergeometric Distribution.
First-Stage and Reduced Form.
- Standard Normal Distribution.
- Multiple Instruments: Two-Stages
- Chi-Squared Distribution.
Least Squares (2SLS) Estimator.
1
, Rodrigo Antón García
o 1 – Introduction. o 2 – Basics.
Stata is a statistical package that includes a • 2.1 – Stata Files Extensions.
wide variety of capabilities, like data
management, statistical and econometric The following extensions are used/created by
analysis, graphics, etc. The user’s interface Stata:
includes the following:
- .dta: datasets that have been created
- Command Window: here we type all by Stata (using the command save).
the instructions (commands) that we ask - .do: programs (sets of instructions)
Stata to execute. Alternatively, a set of lines created with the Stata Do-File Editor.
of commands containing instructions can be - .dct: dictionaries; these are auxiliary
executed from the Do-Editor Window (See in files that are used to import data from text
the Stata Menu/Window/Do-file editor/New files.
do-file editor or just click on the icon - .smcl: log file (explained below) that
displaying a white paper and a pencil). Files can be read by Stata only.
created from the Do-editor (these are called - .log: log file that can be read by any
Stata programs) have the extension .do at the text editor.
end of the filename.
- Results Window: this space is used to • 2.2 – Basic Commands.
display all the results and output originated
by the commands that we have typed. To clear all the variables that Stata has
- Variables Window: shows all the stored in the memory (this is necessary for
variables currently stored in Stata’s memory. importing new data) we can type in the
We can also visualize these variables as in a Command Window:
spreadsheet by typing in the Command
Window browse or br followed by the clear
variables to be displayed (default shows all).
This is the Data Editor window. If we want When we need to learn the use of a command,
to make changes in these data, we will type what options it allows, or to see some
edit in the Command Window. examples of its use, we can type
- Command History: keeps a record of
all the commands used in each session. help name_of_the_command
- Current Working Directory: this
shows the current directory in the file system If we are unsure about the name of a
of your computer from where Stata will read command we are looking for, we can use
or save any files. Stata automatically seeks in search instead.
that directory programs, files, or anything
else it might request. This can be changed by search name_of_the_command
writing cd path_to_the_new_directory or
from the Stata menu: File/Change Working Any command in Stata that is preceded by a
Directory. star character ( ∗ ) will be regarded as
2
