Brief Introduction to STATA 2 Hours

1
Brief Introduction to STATA (2 Hours)
Fabian Waldinger (LSE CEP)
2
Data for this Course

• Download Schooling 1.dta, Schooling 2.dta,
  and Schooling 3.dta from http//personal.lse.ac.
  uk/waldinge/
• and save them on your H Drive
• Open Dataset using the Open Icon from STATA

3
Opening STATA

• Start Menu -gt Programmes -gt Statistics
• -gtSTATA Icon
• 4 Windows
• Command Window
• Results Window
• Variables Window
• Review Window

4
Opening Datasets

• Usually data is not in STATA format (.dta)
• You can copy and paste from Excel
• Use Stat/Transfer program (see handout)

5
Preliminary Commands

• Adjusting memory size
• If you use large datasets you may have to
  increase the memory which the computer reserves
  for STATA
• set mem 30m
• (allocates 30 Megabyte to STATA. Other
  abbreviations
• k kilobyte m megabyte g gigabyte)
• Be careful if you allocate too much memory to
  STATA the processing may become very slow
• Changing the way results are reported in the
  output window
• not making the reporting of results be stopped
  if output window is full (convenient if you use
  log files see below)
• set more off

6
Saving Data

• Click Save Icon
• Saves Dataset as it is at that moment and
  replacing the old version
• (Equivalent to typing
• save H\Schooling 1.dta, replace )
• Preserve and Restore
• convenient for temporary changes to the dataset
  which you do not want to save for later uses. Eg
• preserve
• (photocopies the dataset as it stands)
• (then do all the changes to the data and
  possibly run regressions)
• restore
• (to go back to the originally saved dataset)

7
Examining the Data

• Look at Data in the Data Editor
• Be careful to not make arbitrary changes to the
  data
• Summarize
• Provides summary statistics
• sum
• Tabulate
• Gives frequencies of variables at each value of
  the variable
• (do not use if variable takes too many different
  values)
• tab s

8
Organising the Data

• Rename
• Changing the name of a variable.
• rename lwage logwage
• Recode
• Changes the values of a variable
• recode black 0 .
• (assigns value . (missing value in STATA) if
  smsa was 0 before)
• recode black .0
• Assignment of certain values to variables are
  made using one sign

9
Adding Observations (Append)

• Adding additional rows to the dataset
• Adding additional observations with data on the
  same (or almost the same) variables

If one of the datasets has fewer variables than
the combined dataset will feature all variables
but will have missing values for all observations
from the other dataset append using
H\Schooling 2.dta save H\Schooling 4.dta
10
Adding Variables to Dataset (Merge)

• Adding additional Columns to the dataset
• Adding additional variables the same
  individuals/firms/countries (or almost the same)

A
B
A
B

• In order to apply the merge command you first
  have to sort the data according to the variable
  you are going to use to identify individual
  observations
• Sort Master dataset and save it
• sort id
• 2) Open Use dataset, sort it according to the
  same variable, and save
• sort id

11
Adding Variables to Dataset II

• Reopen Master dataset and then merge using
  dataset with it
• merge using H\Schooling 3.dta
• The Merge command creates a new variable called
  _merge which indicates how well the merging
  process was carried out. Investigate this with
• tab _merge
• The _merge variable can take three values
• 1 observations from the master dataset did not
  match with observations from the using dataset
• 2 observations from the using dataset did not
  match with observations from the master dataset
• 3 observations from both datasets matched

12
Dropping and Keeping Variables

• Before we do this exercise we preserve the data
• preserve
• Keep
• Enables you just to keep the variables specified
  which you want to use for your analysis.
• keep s age e
• Drop
• Drops the variables which you specify in the
  variable list.
• drop

13
Dropping and Keeping Variables II

• Can be combined with relational operators or
  logical operators
• Relational Operators Logical Operators
• equal to and
• ! not equal to or
• gt greater than ! not
• gt greater than or equal to
• lt less than
• lt less than or equal to
• drop if (sgt12 s lt 8)
• restore

14
Creating New Variables

• Generate
• Generates a new variable
• gen college 1 if sgt12
• gen e2 ee
• Egen
• typically creates variables based on summary
  statistics
• egen meanwage mean(logwage)
• Replace
• Modifies existing variables in exactly the same
  way as gen creates new variables
• replace college 0 if college.

15
Log-Files and Do-Files

• Log-Files
• Record all commands and the output of your STATA
  session.
• Do-Files
• Combine many commands in a file which can be
  saved to make changes and rerun it later on
• Open Do-File Window and type
• log using H\schooling.log
• sum
• (other commands)
• log close

16
OLS Regression

• Running an OLS regression
• first specify the dependent variable and then
  all the explanatory variables
• reg logwage s e e2 black
• Can be combined with logical operators
• reg logwage s e e2 black if college1

17
IV Regression

• Running an IV regression
• Instrumenting schooling with the dummy whether
  you live near a four year college
• ivreg logwage (snearcoll) e e2 black
• First Stage is not reported. You would have to
  estimate in the standard way using OLS if you
  want to investigate it

18
Getting Information on STATA

• STATA help
• just type help followed by the command you want
  to use. E.g.
• help reg
• STATA Manuals
• Available in the Computer Manual Section of the
  Library
• Notes on EC 402 webpage

19
Some Useful Links

• http//rlab.lse.ac.uk/data/stata_info/
• http//rlab.lse.ac.uk/it/STATA.asp
• http//ittraining.lse.ac.uk/documentation/Files/St
  ata-805-Get-started-with.pdf
• http//econ.lse.ac.uk/staff/amanning/courses/ec402
  /statanotes.pdf
• http//www.stata.com/links/resources1.html