stata class

8/2/2019 Stata Class

1/22

STATA LESSON

Applied Economics

March 2012

Instructor: Ainhoa Aparicio-Fenoll

Outline:

1. Introduction

2. Dataset management

3. Generating and recoding arrays

4. Control flows

5. Data overview

6. Stata for econometrics

Materials:

http://intranet.barcelonagse.eu

http://www.stata.com/links/resources1.html
http://intranet.barcelonagse.eu/http://www.stata.com/links/resources1.htmlhttp://intranet.barcelonagse.eu/http://www.stata.com/links/resources1.html


2/22

1. INTRODUCTION

STATA is a general command-drive package for statistical analysis, data management and

graphics. It can be considered a stat package, like SAS, SPSS, RATS, or eViews. STATA

operates in a graphical (windowed) environment.

Basic features

When to use STATA? STATA has three major strengths, namely, data manipulation,

statistics and graphics. Which STATA to use? There are different versions in terms of maximum number of

variables and matrix size.

In general, Small STATA


3/22

Interface description

STATA, by default opens with four windows:

Results - where all of your commands and their results are displayed (with the exception

of graphs which are displayed in their own window). Anything displayed in blue can be

clicked on to get help or other information. When results are too long it shows the word

more. You should click on it to continue watching the results. If you want all the results

to be displayed even if you miss the upper part, just type: set more off, permanently Review - just your commands are displayed here. You can click on any command in the

window and it will be pasted to the command window. The Review window has one extra

option in its windows icon menu: "Save Review Contents." This will allow you to save

everything in the review window to a file for later use. Although this is not a substitute for

using log and do files. Command - this is where you type your commands when working in interactive mode.

Any command typed here would be executed just by pressing the enter key. Everything

you type in here is echoed in the Results window as well as the Review window. The

"Page Up" and "Page Down" keys can be used to scroll back and forth through

Result window

Command windowReview

window

Variabels

window
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


4/22

commands you have executed previously. You can also copy and paste between this

window and your "do" file. Variables - a list of all your variables and their labels is displayed here. You can click on a

variable here and it will be pasted to the command window.

There are three other windows in STATA:

Do-file editor - this is STATA's simple text editor for writing do-files, or programs. You

should do all of your work in a do-file so you can reproduce what you did later on. This is

accessible through Window>Do-file editor>New Do-file or by clicking on the envelope

icon. You can execute all commands or just some of them and you may choose the

results to be displayed on the results window or not. Viewer - the viewer is used for displaying help and log files. Like the Results window,

anything displayed in blue can be clicked on for more information. This window is reachedby clicking on Help>Content or by typing help in the command window.

Graph - as you may have guessed, this is where all of your graphs will be displayed.

The tool bar:

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

(1): Opens a dataset

(2): Saves a dataset

(3): Prints graphs or contents of viewers

(4): Begins, closes or suspends a log-file(5): Brings viewer in front

(6): Brings the result window in front

(7): Opens a do-file

(8): Opens the data editor

(9): Opens the data browser

(10): Breaks/stops the operation

M enu bar:

File: To open, save, view, launch do-files, save graphs, print graphs or results
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


5/22


6/22


7/22

You can add more results to an already closed log by typing: log using

/logfilename.log , append

Or you can simply replace an existing log with the command: log using

/logfilename.log , replace To take a look at a log-file go to File>Log>View and browse it.

Do-file

The do-file is an ASCII file that collects a collection of commands to be executed sequentially. It is

very useful because it makes it very easy for everybody to replicate your results.

When starting a do-file, you should specify:

That the old data should be removed from the memory, otherwise you would not be ableto upload your dataset. Use the command clear

The directory where you will be working (that directory should contain your data,

dictionary, etc.). The corresponding command is cd C:/EXAMPLE

The amount of memory to be allocated to STATA. You must make sufficient memory

available to STATA to load the entire le, since STATAs speed is largely derived from

holding the entire data set in memory. Use the command set memory

Alternatively you may want to set the maximum number of variables ( set maxvar ), the

maximum matrix size ( set matsize ) and set the more option off ( set more off ). Now you can upload your new dataset and work on it.

If you want to see the results of the operations in the do-file displayed in the results window, type

or click the icon for run dofilename and if you do not want to see the results in the do-file

displayed in the results window, type or click the icon for do dofilename.

If you want to execute only some of the commands in the do-file, select them and type or click on

the icon for run dofilename or do dofilename.

(1)(2) (3)(4) (5)(6) (7)(8)
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


8/22

(1): New do-file

(2): Open a folder where you can find an old do-file

(3): Save the do-file

(4): Print the do-file(5): Find the do-file

(6): Un-do

(7): Run the commands of the do-file

(8): Do the commands of the do-file

(9): Search

2. DATASET MANAGEMENT

Getting data into STATA

You can input data by hand:

set obs

edit

Type the data in the bar above and press enter.

Or alternatively,

input v1 v2.vn And typing the numbers on the command window

end

Note: For the missing values, type .

There are several commands to input already created datasets. You will need to use one

or another depending on the format in which the dataset is provided.

- insheet : This is used to input ASCII files where the file is delimited by tabulations or commas. For this command to work the data must contain only one observation per line.

For instance, you may type: insheet using mydata.txt. You may even want to specify

which the delimiter is, then type: insheet using mydata.txt, delimiter (;)

- infix : You may need this command if the variables in your data are not in a fixed format.You should declare their position by typing infix v1 1-4 v2 6-7 v3 9-11

- infile: It is useful when inputting complex ASCII files where several observations appear inthe same line or where the variables are separated by tabulations, commas or spaces.

For instance, you may type infile var1 var2varn using filename.txt , automatic
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


9/22


10/22

Joining, expanding and collapsing datasets

Joining datasets

1. Merge joins corresponding observations from the dataset currently in memory (called the

master dataset) with those from the STATA-format dataset stored as filename (called the using

dataset) into single observations.

Sintax:

merge [varlist] using filename [, keep(varlist) unique uniqmaster uniqusing nolabel update

replace nokeep _merge(varname) ]

After each merge a new variable is created (the default is _merge) whose values are:

_merge==1 obs. from master data

_merge==2 obs. from only one using dataset

_merge==3 obs. from at least two datasets, master or using

2. Append appends a Stata-format dataset stored on disk to the end of the dataset in memory.

Sintax:

append using filename [, nolabel keep(varlist) ]

Expanding and colla psing datasets

1. collapse [(stat)] varlist [ [(stat)] ... ] [if] [in] [weight] [, options] It transforms the database in

memory into a database of sums, means, medians, etc. Example: "collapse (sum) wage, by(zone

education age)"

2. expand [=]exp [if] [in] Example: "expand 2" / "expand pop. It multiplies the number of lines a

certain number of times (2 in the first example) or according to the value of one variable in thatline (pop in the second example).

3. GENERATING AND RECODING ARRAYS

Variables

generate varname=x to create a variable

replace varname=x to replace the value of a variable

label variable varname to give a variable a certain description
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


11/22

Note: these commands follow the common syntax

[by varlist:] command [ varlist] [if exp] [ in range ], [ options ] and thus can be applied to a restricted

sample or as an extended version.

Where:

by varlist:

It is used to group data (before you have to sort the data according to the varlist by the command

sort varlist)

c ommand:

You can find a list of commands below

in:

It is used to restrict the sample to a certain range (e.g. in 4, restricts to case 4, in 1/4, which

shows cases 1 to 4)

Options:

Various options are possible depending on the command (see help)

e.g. ,replace replaces a variable//dataset etc. in case it already exists

if :

You can restrict the sample by using the following expressions:

Arithmetic Logical (numeric and string)+ addition ~ not > greater than- subtraction | or < less than* multiplication & and >= greater or equal/ division 0, 0 if x=0 and -1 if x


12/22

Furthermore you might want to use some system variables, here a selection:

Command System variable _n Index of current observation _N Total number of observations _all All variables _b Vector of regression coefficients

Vector of standard errors of regression

coefficients

Note: Depending on the mathematical/statistical function you want to use you might have to use

an extension of generate called egen .

egen varname=function(varx) is used to create a variable which is the mean, median,of

another variable

Below a selection of functions where you have to use egen :

Command Mathematical & Statistical Functionmean (varx) Mean of variable xMedian (varx) Median of variable xsum (varx) Sum of variable x (Note: gen varz=sum(varx)

varz= cumulative sum)

rank (varx) Rank of variable x, highest value gets value1count (varx) The number of nonmissing observations of x

Examples of discrete variables

"generate byte rich=(wage>100)": It generates a dummy variable.

"generate byte rich=(wage>100) if !missing wage": It generates missing values for rich if wage is

missing.

"generate young=inrange(age,18,25)". Young takes value one if age is between 18 and 25.

"generate cs=2 if inlist(csp,1,2,3,4)". The variable cs takes value 2 if csp takes any of the listed

values.

"tabulate csp, gen(csp)". This generates as many dummy variables as values are taken by csp.

"generate size=recode(nb,0,5,10,15)". This command generates a new variable size that takes

the values 0, 5, 10 and 15 whenever nb is zero, between 1 and 5, between 6 and 10 and 11 or

more, respectively.

"separate treff, by(csp)". This generates as many variables as values takes csp. Each variable

is equal to csp for a certain value and zero otherwise.

"gen szone=group(size zone)" is used for generating identifiers. It gives a code to any

combination of the values of size and zone.
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


13/22

"egen sizegr=cut(size), group(5)" It divides the variable size in 5 different categories with

homogenous frequencies.

"egen sizegr=cut(size), at(5,10,15)". It divides size into two categories, one from 5 to 9, and

another from 10 to 15.

"egen delta=diff(v1 v2 v3)". Delta is equal to one only if the three variables take the same value.

Examples of continuous variables

"gen suma=sum(wage)". This gives the cumulative sum of the variable wage. This is different

form "egen suma=sum(wage)" which generates the total sum.

"generate id=_n". It gives the values 1, 2, 3,... to each observation following the order in which

they appear in the data.

"(bysort year:) egen avg=mean(wage)"/ "egen mini=min(wage)" / "egen maxi=max(wage)" /

"egen sd=std(wage)" / "egen moda=mode(wage)". It generates a variable equal to this statistic for

all observations (or observations in the same group in the case of using bysort).

"egen avgr=rowmean(wage1 wage2 wage3)" calculates the mean of the specified variables for

each observation.

"egen sumr=rowtotal(wage1 wage2 wage3)" calculates the sum of the specified variables for

each observation.

Examples of string variables

"generate str4 zone4=substr(zone,1,4)". It takes the 4 first digits of the variable zone. Note:

zone must be a string.

"egen abc=concat(a b c)". It generates one variable that results from concatenating the three

variables a, b and c.

Variable elimination

Once we have created the variables we might want to get rid of some of them. These we caneither do by eliminating the variables that we are not interested anymore (var1 varn)

drop var2 var1 varn

or by keeping the ones that we still want to use (var1 var2 varn)

keep var1 var2 varn

Furthermore we can change the name of a variable (from oldname to newname)

rename oldname newname
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


14/22

Examples of data elimination

"rename sex gender": It renames the variable sex, gender.

"replace zone=4 if zone==2 | zone==3". It will write 4 whenever the variable of zone was 2 or 3.

"recode zone (2 3=4) (6 7=5), gen(zone2)". This way you generate a new variable with values 4and 5 according to whether the values of zone were 2 / 3 or 6 / 7.

Macros

Macros store information as strings. You can collect several variables under a common

macroname.

global macroglobalname var1 var2 varn

local macrolocalname var1 var2 varn

Both macros are working exactly the same way (you assign strings to specified macro names and

can call them later by calling it `macroname). The main difference is that a global macro may be

used by any program whereas a local macro is only for private use of the program in which you

define it.

When you want to use the variables contained in a macro, you should use the following syntax:

sum $macroglobalname

sum `macrolocalname (be careful with the commas)

Scalars

The scalars are variables that contain one single element. To handle them, you must use an

specific syntax:

- To define the content of the scalar:

scalar a=normal(0.7)

scalar b=199876

scalar c=V[1,3]

- To list the contents of scalars:

scalar list

scalar list a b
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


15/22


16/22

}

else {

command

}

The if command evaluates expression1 (e.g. var1>0). If the result is true, then it executes the

command inside the first set of brackets. If not, it evaluates expression2 (e.g. vr1


17/22

5. DATA OVERVIEW

Descriptive statistics

describe

It provides basic information about the file and the variables (number of observations, number and

names of the variables, format of the variables, labels of variables if attached)

browse

It opens the data-browser where you can see the dataset

list varlist

It diplays all the data about the listed variables on the screen

Summarize varlist

It provides a summary statistic about the variables listed (frequencies, mean, min, max).

Note: a possible option is , detail which provides more details such as percentiles, kurtosis etc.

tabulate varlist

It shows all different values of the variables with their frequencies

table varlist

It displays higher dimensional tables (see help for options)

correlate varlist

It gives the correlation coefficient between the variables named

Graphs

An easy way to create graphs is to use the graph option in the menu. There you can find the

whole variety of graphs offered by STATA.
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


18/22

There are two different graph commands in STATA:

graph plottype [var1] [var2]

graph twoway plottype [var1] [var2]

The first command draws (in the announced plot-type) a one dimensional graph of the means of

the variable listed. The second graph command shows the two-dimensional relationship between

the variables listed. [var1] appears on the y-axis, and [var2] on the x-axis of the graph. The default

extension for Stata graph files is .gph which Stata will automatically add to the file name. You can

select the font, colors, line thickness and other options by selecting Graph Preferences in the

Prefs menu.

The following table gives an overview of the type of graphs you can draw in STATA

Graphics* Descriptiontwoway[plottype] vary varx

plottype : scatter, line, bar etc.

2-dimensional family of plots, all of which fit on

numeric y and x scalesgraph bar (mean) vary , over (varx) vertical bar charts (y axis is numerical, and the

x axis is categoricalgraph dot (mean) vary , over (varx) horizontal dot charts (categorical axis is

presented vertically, the numerical axis is

presented horizontally)graph pie vary , over (varx) Pie chartsgraph matrix matrix plot of two-way scattergramsgraph save graphname saves a graph under graphname with the

default extension .gphgraph use graphname.gph redisplays the graph graphnamegraph combine graphname1 graphname2 combines several graphs into one

Example of graphs

graph hbar (asis) pop, over(voter) title("Population by city") ytitle("Number of inhabitants")

note("The source of the data is unknown")
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


19/22

0 50,000 100000 150000 200000Number of inhabitants

15

1413121110

9876543

21

The source of the data is unknown

Population by city

twoway (scatter price mpg, mlabel(mpg) xline(20) xscale(range(10(10)45))) (lfit price mpg),

legend(off) title("Price over mileage") ytitle("Mileage") xtitle("Price") graphregion(fcolor(green)

lcolor(green))
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


20/22


21/22

-------------+------------------------------ F( 3, 236) = 1.08

Model | .728611679 3 .24287056 Prob > F = 0.3571

Residual | 52.9338883 236 .224296137 R-squared = 0.0136

-------------+------------------------------ Adj R-squared = 0.0010

Total | 53.6625 239 .224529289 Root MSE = .4736

------------------------------------------------------------------------------

male | Coef. Std. Err. t P>|t| [95% Conf. Interval]

-------------+----------------------------------------------------------------

age | -.0153606 .0162436 -0.95 0.345 -.0473615 .0166404

grade | -.030674 .0201378 -1.52 0.129 -.0703469 .0089989

toefl | -.0003715 .0013527 -0.27 0.784 -.0030364 .0022934

_cons | 1.372898 .5609794 2.45 0.015 .267731 2.478065

Anova block: This block shows the sum of squares for both the explained part of

the model and the residuals. df shows the degree of freedom and

MS the mean squared error with respect to the degrees of freedom

Modelfit block: This block shows the number of investigated units, the F-Statistic and the

adjusted R squared value.

Coefficient block: This block provides the estimation results for all control variables, incl.

the coefficients, the standard errors, the value of the t-statistic and the p-

value as well as the confidence interval.

For further advanced estimation procedures the following commands might be helpful:

Regression procedures DescriptionRreg Robust regressionIvreg Instrumental variables regressionHeckman Heckmans selection model

probit/logit Probit/Logit analysismlogit Multinomial logitTobit Censored-normal and Tobit regression

Arima* Autoregressive integrated MA models Arch* AR conditional heteroscedasiticity estimators xt*... Panel analysisst* Survival time data* data must be time-series/panel data/ survival data; use the command tsset panelvar timevar or

stset.

You may want to type quietly in front of the regression command to avoid all the table being

displayed. Specially, if you just want to make predictions or you are just interested in the

coefficients.

Post-estimation commands
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html


22/22

After having done a regression you might want to use the results for further analysis, e.g. some

significance test or graphical analysis. In the following table you can find some useful commands

to save regression results, to calculate marginal effects or to do certain types of tests. For more

detailed information see the HELP function.

Post estimation command Description predict name, options saves predictions (xb), residuals (res), influence statistics, etc. predictnl name, options point estimates, standard errors, testing, and inference for

generalized predictionsmfx marginal effects or elasticities in a nonlinear equation (e.g. probit,

logit etc.)estimate store saves estimation resultsEstat AIC, BIC, VCE, and estimation sample summarymat name = e() saves the estimation results such as the coefficients (e(b)),

variance covariance matrix (e(V)Test Wald tests for simple and composite linear hypothesestestnl Wald tests of nonlinear hypotheseslrtest Likelihood-ratio test (to conduct the test, both the unrestricted and

the restricted models must be fitted using the ML method)hausman Hausman's specification test: tests if an efficient and a consistent

estimator are significantly different.
http://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.htmlhttp://www.stata.com/links/resources1.html

stata class

Documents