Stata Tutorial

Introduction
-1- Environment
-2.a- System Syntax
-2.b- Statistical Syntax
-3- Output
Stata Tutorial IT2010

Francesco Andreoli Andrea Bonfatti
Universit di Verona
January 11-15, 2010 Alba di Canazei
C.I.D.E.
January 11-15, 2010 1 / 59
Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Acknowledgements Department of Economics (DSE) - Universit di Verona --Nicola Tommasi (CIDE and Universit di Verona) provided valuable support with related documents and an early version of this presentation.
C.I.D.E.
January 11-15, 2010 2 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Organization of the Directory

../stata_tutorial | |--> /read_me_first.txt | |--> /tutorial.pdf | |--> /data_files | | | |--> /smallPSELL.dta | |--> /smallPSELL.csv | |--> /smallPSELL.txt | |--> /smallPSELL2.dta | |--> /smallPSELL2.csv | |--> /smallPSELL2.txt | |--> /estimates.txt | |--> /do_files | |--> /do_first.do |--> /do_first.log |--> /model.txt |--> /model.gph
* You are here! *
C.I.D.E.
January 11-15, 2010 3 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Motivation
This tutorial addresses to a beginner level or early trained audience which needs notions as well as operational hints to get started with Stata software. This handout is meant to provide a support for a two hours tutorial, and it can be considered complementary to other more exhaustive and rigorous sources. You are invited to take a look at the Ocial Stata website, http://www.stata.com/bookstore/documentation.html, where you can nd all published documents. The Getting Started with Stata manual is an introductory (although very complete) users manual.
C.I.D.E.
January 11-15, 2010 4 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Motivation
Additional material can be found in ocial websites:
http://www.stata.com/bookstore/pdf/gsw_samplesession.pdf http://www.stata.com/bookstore/pdf/r_intro.pdf (commands) http://www.stata.com/bookstore/pdf/g_graph_intro.pdf (graphs) http://www.stata.com/bookstore/pdf/d_merge.pdf (merge)
and unocial ones: http://www.ats.ucla.edu/stat/stata/ http://www.nyu.edu/its/statistics/Docs/Intro_stata5.pdf http://www.eui.eu/Personal/Researchers/decio/PS/Stata.pdf http://leuven.economists.nl/stata/stataintro.pdf

C.I.D.E.
January 11-15, 2010 5 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Organization of the Tutorial

The tutorial focuses on Stata for Windows package, SE version, actually at the 11th release (the same procedure holds for older releases and other versions). Firstly, the tutorial will exploit the general settings of the environment where the data are stored, managed and analyzed. Here we will look closely on how organize data into folders and how to work with a hierarchy of folders, in order to make the research work intelligible by all audienced. Order is necessary to work scientically, i.e. to be able to replicate an experiment starting from the observed phenomena, preserving the measurability. Moreover, order is a prerequisite for understanding and interpretability of results. On a second stage, attention will be paid to programming and coding and the general setting of the program will be introduced: windows, interface, .do .dta .log .ado extensions, input/output of data les, basic syntax (data management, statistics, regression, principles of graphs). C.I.D.E.
January 11-15, 2010 6 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Organization of the Tutorial

The concluding stage of the tutorial aims at showing how Stata works in practice, by applying the program (.do le) to a small subsample (50 times 8) from the PSELL (Panel Socio-conomique Liewen zu Ltzebuerg) database (.dta le). Finally, Stata output will be described and analyzed (.log le). The handout structure follows the tutorial sectioning. As a general purpose, you will be given with the means to autonomously search and apply new syntax already stored in Stata memory (.ado les) or downloadable from the web sites of Stata Journal, Stata Technical Bullettin or from Repec Library. The use of any particular syntax, as the one treated in classes to perform income distribution analysis, depends on your research scopes. For more advanced users, Stata oers the possibility to code new functions according to program language standards. C.I.D.E.
January 11-15, 2010 7 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
A First Approach...
Open Stata by clicking on the Stata icon. You recognize four windows: a) the output window, reporting code input and associated output from a database in use; b) an input window, where each code, line by line, can be written; c) a review window in which is recorded the code inputted in Stata and d) a variables window, which reports all the variables in use, a sort of summary of the database. All inputs you select can either be written (see point b)) or selected interactively in by the option bar of the program, and are systematically recorded by Stata in the review windows (see point c)). These inputs may refer to statistical operations as means, frequency tables, regressions, etc., as well as changes to the database, transformations or the creation of new variables or modication of the entire database. In the former case, the output (the value of the mean, the table, the regression coecients, etc.,) is displayed in the output window (point a)), while in the latter, the variable window (point d) will report the resulting modications. C.I.D.E.
January 11-15, 2010 8 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
A First Approach...
Input data are stored in the RAM memory of your computer (which is then required to be at least as large as the database dimension) as an n observations by k variables matrix which you do not need to see while programming. In the philosophy of the software, the original database is sacred. For the sake of reproducibility of the experiment you are running on your data, the primitive source of information (i.e. your original database) must be preserved intact and unaected by additional elaborations which may induce errors in future users elaborations, tests or verications. For this reason, any change of the database remains stored only in the RAM memory of your computer and it will be completely deleted when Stata is closed, if you refuse to save changes and results when asked. When you decide to save, it is important to choose what to save and how to save it appropriately. C.I.D.E.
January 11-15, 2010 9 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
A First Approach...
Three fundamental steps:
1
If the original database has been modied, a new database must be created containing all new variables and transformations performed. In this way, the original database will be preserved as an independent object from the new one, but any information about how to go from one database to the other will be lost. If the original database is not modied, there is no need to save it and you will not be even asked to do so. Stata opens and saves databases in the .dta extention. Output results can be saved in a log-on le, reporting a list of your results in a .txt extension. From this le you can copy tables or coecient results to be used in you research report. In Stata jargon, this is a log-le. Inputs as well can be saved, reporting the full list of commands appearing in the review window on a .txt document. This list can be used by other readers to understand how the new database and the output saved in the log le were obtained. In Stata jargon, this is a do-le. C.I.D.E.
January 11-15, 2010 10 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Setting the Environment

The scope of this section is to show how to correctly manage your input, commands and output les for the sake of reproducibility and intelligibility of your work. The rst object you need to manage is the initial source of data. The original database is your primitive source of information, therefore it must be preserved on its original status. Data can be read, and you can work with them, but never modify or delete them because they represent the primitive source of information. Stata programming must be created to work with relative paths which can be used in any le system. A path, the general form of a lename or of a directory name, species a unique location in a le system. A relative path is a path relative to the working directory of the user or application (eg: ../stata_tutorial/do_files/do_first.do) which does not need to depend on the full path root usually system specic (C://Francesco_Andrea/IT2010/stata_tutorial/do_files/do_first.do). C.I.D.E.
January 11-15, 2010 11 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Setting the Environment

Golden rules in Stata Generate a directory projects in which to put in all your current projects; Inside a specic project, take documents, data, programs as distinct elements; In the data directory put your database. It is your original object; In the programs directory put your coding and the statistical results; Statistical results can be used either directly in a paper (graphs, tables,...) or by other softwares (derived databases, parameters estimates,...); Use simple names for les. File log in accordance with programming le. Be sequential. Use comments and always describe what you are doing. C.I.D.E.
January 11-15, 2010 12 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Files Extentions in Stata
.dta data les, also in .csv or .txt .do program le containing coding, the starting point .log; .smcl output logging les, results storing .ado les used by programmers, advanced code .mata le extention in Mata environment .hlp; .sthlp help les .scheme; .style; .gph extension for graph attributes and to save graphs
C.I.D.E.
Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial January 11-15, 2010 13 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Commands
All Stata commands are contructed following a precise syntax structure, allowing you to recognize always what is the command employed, the variables in use and the options, even for previously unseen objects. In this way, you can search for command help les. The general syntax of a command:
command varlist =exp if in weight , options
mandatory coding between brackets optional coding between commands in { } are parameters whose value must be specied underlined letters are abbreviations for commands parts in , are the commands options
C.I.D.E.
January 11-15, 2010 14 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Help
The general syntax for help, search, find
help command_or_topic_name , options search word word ... , search_options findit word word ...
Help is the most important command in Stata. It provides the full dictionary and help of the commandlist specied. Use help! You cannot (and you must not) recall all commands options and possible applications. Each help le, appearing on Stata screen, has the form:
1 2 3 4 5 6
Command syntax Description Options Additional options for related commands Examples See Also
C.I.D.E.
January 11-15, 2010 15 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Prepare Your .do File

Stata works with commands, so you need to know the syntax Coding in .do les must be intellegible: use comments but be sequential and schematic Do le is the source of your commands. In case of errors, you can easily correct it an re-run the program. This would replace erroneous estimation previously done with new ones Follow a operational sequence:
1 2 3 4 5
double click on .do le: positioning Stata write and run the .do le by Stata do-editor (or parts of it) look at results on .log les save changes and a new database if the original one has changed keep all the results in a organized directory C.I.D.E.
January 11-15, 2010 16 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Prepare Your .do File

In any application of a dole.do, you are recommanded to follow the three steps here presented (; can be deleted if no delimit is specied): .do le
#delimit; set more off; clear; set mem 150m; capture log close; log using dofile.log, rep; ... command list; ... exit;
Stata runs Stata
.log le Output is recorded in dole.log as it appears in result window dole
do dole
C.I.D.E.
January 11-15, 2010 17 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Syntax: an Introduction
In the following 2 sections we will describe mainly used syntax in Stata programming. Common Syntax section refers to commands and options which allow to use other statistical commands or to menage the data stored in memory. They are mainly functional to a statistical program. Statistical Syntax section contains mainly used statistical tools in Stata to perform data management, data analysis, graphic analysis, regression and testing. It is worth note that our list does not exhaust the full set of statistical routines in Stata. Many of them can be derived (see the help les) from the ones we put here, while other can be found in Repec Lybrary or looking at the Stata online help. All these commands apply exclusively to data stored in Stata memory at the moment in which they are used, and they provide a syntetic output (coecients, estimates) as well as new variables to be added at the database. We will see in the next section how to work with estimates. C.I.D.E.
January 11-15, 2010 18 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Change Folder
Shows current working folder
pwd
Create a new folder

mkdir path directoryname
Folder content
dir path directoryname
Change working folder

cd path C.I.D.E.
January 11-15, 2010 19 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Search and Upload New Commands

Search
ssc hot , n(#) ssc new
Update the software

update all
Update commands
adoupdate pkglist , options adoupdate, update
Note: These commands work i Stata is connected with the internet.

C.I.D.E.
January 11-15, 2010 20 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Use .dta Files

Set memory capacity
set memory # b|k|m|g , permanently
Use data /1
use lename , clear
Use data /2
use varlist if in using lename , clear nolabel
Compress the data

compress varlist C.I.D.E.
January 11-15, 2010 21 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Use Delimited Format Data

Each variable realization is separated from the other by a given separating character or by tabulation Delimiters: , insheet
insheet varlist using lename , options
<space>
<tab>
Between the most important features: tab: to indicate that data are devided by tabs,.txt comma: to indicate that data are devided by commas, .csv delimiter: species the delimiting object between quotations clear: to clean other data stored in memory
C.I.D.E.
January 11-15, 2010 22 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Use Not Delimited Format Data

- Each varible is identied depending on the required space, in a .txt data le - Starting from the left, position is determined in terms of columns, while each observation is a row - A dictionary set boundary limits of each variable
inx
infix using dictlename if in , using(lename2) clear
Build a dictionary le
infix dictionary using datafile.ext { var1 s1-e1 var2 s2-e2 var3 s3-e3 } C.I.D.E.
January 11-15, 2010 23 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Export Data
In Stata format save & saveold
save lename lename , replace , replace
saveold
In text format outsheet

outsheet varlist using filename if in , options
comma data separated by , (usually .txt) instead of tabulation (.csv) delimiter(char) other delimiter, for instance ; nolabel export the numeric value, not the label replace overwrite the existing le
C.I.D.E.
January 11-15, 2010 24 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Key Variable(s)
Denition A key variable(s) is that variable (or set of variables) which uniquely identies each observation How to identify the key variable duplicates report
duplicates report varlist if in
C.I.D.E.
January 11-15, 2010 25 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Qualiers in and if
in restricts the set of observations to which a command applies it refers to the rows identifying the observations not applicable to all commands not sensitive to the sorting of data if species the conditions for the execution of a command it applies to the values of variables and always refers to observations not applicable to all commands not sensitive to the sorting of data it requires relational qualiers
C.I.D.E.
January 11-15, 2010 26 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Relational-Logical-Jolly Operators
Relational operators > strictly greater of < strictly less of >= greater or equal to <= less or equal to == equal to (note the use of the double sign ==) = or != dierent from Logical operators & (and) it requires that both relations hold | (or) it requires that at least one of the relations holds Jolly characters * any character and for whatever number of times ? any character for one time only - a contiguous series of variables. (Note, this espression depends on the order of variables!)
January 11-15, 2010
C.I.D.E.
27 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
by and bysort
by repeats the command for each group of observations for which the values of the variables in varlistare the same. Without the sort option by requires that the data be sorted by varlist bysort performs the sorting of varlistand then repeats the command by and bysort
by varlist: command command
bysort varlist:
Not all commands are byable, that is support bysort

C.I.D.E.
Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial January 11-15, 2010 28 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Describe and Label

describe
describe varlist , memory_options
memory_options short less information and memory space allocated, number of variables, number of observations detail more detailed information fullnames variable names not abbreviated
codebook
codebook varlist if in , options
notes displays the notes associated to the variables tabulate(#) shows the values of categorical variables problems detail reports problems to the dataset (missing variables, variables without label, constants) compact yields a more concise report on variables
C.I.D.E.
January 11-15, 2010 29 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Describe and Label

Put a label to your variable (var# by default)
label variable varname label
Dene a label (a)...

label define label_name #1 desc 1 add modify nofix #2 desc 2 ...#n desc n ,
... and label your values (b)

label values varname label_name , options
C.I.D.E.
January 11-15, 2010 30 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Rename Variables
Put a new name on variables
rename old_varname new_varname renvars renvars varlist \ newvarlist , display test varlist , transformation_option , display test
symbol(string)
display displays each change upper convert the names in upper case lower convert the names in lower case prefix(str) assign the prex str to the name postfix(str) add str at the end of the name subst(str1 str2) replace all str1 with str2 (str2 can be empty) trim(#) take only the rst # characters of the name trimend(#) take only the last # characters of the name
C.I.D.E.
January 11-15, 2010 31 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Modify Data
Your original database (n k) can be integrated, compressed or shaped: Add observations: type help append You add observations to your database from other data sources (m observations) obtaining a new database (n + m) k with m > 0 and k = k required to have a balanced sample (otherwise missing values generated for surplus variables). Add variables: type help merge or help mmerge You add variables to your database from other data sources (h variables) obtaining a new database n (k + h) with h > 0 and n = n required to have a balanced sample (no missing observations). A variable _merge {1, 2, 3} is created, showing if missing observations result from merging. Both databases used must have the same key variable(s). C.I.D.E.
January 11-15, 2010 32 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Modify Data
Transform and preserve information: type help reshape Transform a (n m) (k h) database in a n (k h m) format (wide option) or in a (n m h) k format (long option). It is not required that all n groups display m observations. Transform but not preserve information: type help collapse Transform a (n m) k database in a m k format, where k k contains statistics of k as mean, sd, count, freq,.... You loose information but you can work with subsample group averaged data. Note that information lost cannot be restored from the last database saved.
C.I.D.E.
January 11-15, 2010 33 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Sort, Keep of Drop Variables/Observations

Order variables
order varlist move varname1 varname2

Sort observations
sort varlist in gsort +|-
, stable +|varname ... , options
varname
Keep or drop observations
keep if condition drop if condition sample # if in , count by(groupvars)
Keep or drop variables
keep (or drop) varlist
C.I.D.E.
January 11-15, 2010 34 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Create Variables
These commands allow to operate with numeric variables only. They are column operators which return a new column of values in the database. generate
generate
type
newvarname=exp
if
in
An algebraic function between existing variables abs(x) generate the absolute value of each value of the variable x int(x) returns the integer obtained by truncating x toward 0 ln(x) returns the natural logarithm of x max(x1,x2,...,xn) returns the maximum value of x1, x2, ..., xn min(x1,x2,...,xn) returns the minimum value of x1, x2, ..., xn sum(x) returns the running sum of x treating missing values as zero uniform() returns uniformly distributed pseudorandom numbers on the interval [0,1) invnormal() returns the inverse cumulative standard normal distribution lower(s), upper(s) return s in lower (upper) case letters
C.I.D.E.
January 11-15, 2010 35 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Create Variables
Advanced generate command (for column-wide functions)
egen type newvarname = fcn(arguments) if in , options
count(exp) creates a constant (within varlist) containing the number of nonmissing observations of exp mean(varlist) creates a constant (within varlist) containing the mean of exp rowtotal(varlist) creates the (row) sum of the variables in varlist, treating missing as 0 group(varlist) creates one variable taking on values 1, 2, ... for the groups formed by varlist Replace values of a variable
replace varname =exp if in C.I.D.E.
January 11-15, 2010 36 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Create Variables: Dummy Variables

Dummy variables: variables taking on the values (1), when the character of interest is present, or (0) otherwise. To generate the variable you can either use generate and than replace missing values generated; or: 1) Recode your data into a dummy variable
recode varlist (erule) (erule) ...

generate(newvar) create a new variable
if
in
, options
prefix(string) create new variables with the prex string This command can be used simply to change sequences of values 2) Follow a programming procedure
char varname omit value xi , prefix(string) : term(s) C.I.D.E.

37 / 59
char species the reference variable of a set of dummies (to evitate perfect collinearity) term species with a i.varname the variables that must be converted in dummies. January 11-15, 2010
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Continuous Variables
To obtain statistics as output coecients
summarize
varlist
if
in
weight
, detail
To obtain statistics between data
fsum
varlist
weight
if
in
, options
where the main option is stats() with these possibilities: n, miss, abspct, mean, vari, sd, se, p1, p5, p25, p50 or median, p75, p95, p99, min, max To obtain statistics on the mean (like ci and se)
ci
varlist
if
in
weight
, options
Percentiles on variables: generate pctile values or ranking function
pctile
type
newvar = exp if in
if
in
weight
, options C.I.D.E.
January 11-15, 2010 38 / 59
xtile newvar = exp
weight
, options
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Continuous Variables
Compute the correlation (1)
correlate
varlist
if
in
weight
, correlate_options
Compute the correlation (2)
pwcorr
varlist
if
in
weight
, pwcorr_options
obs print the number of observations for each couple of variables sig print the signicance level of the correlation star(#) display with the sign * signicance levels less than # bonferroni use Bonferroni-adjusted signicance level sidak use Sidak-adjusted signicance level
Check for outliers
hadimvo varlist if grubbs varlist if
in , generate(newvar1 newvar2 ) p(#) in , options C.I.D.E.

January 11-15, 2010 39 / 59
drop eliminate the observations identied as outliers generate(newvar1 ...) generate dummy variables for identifying outliers
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Discrete Variables
Table of frequences for single variable(s)
tabulate varname if tab1 varlist if in
in
weight
, tabulate_options
weight
, tab1_options
missing include missing values nolabel display numeric codes rather than value labels sort display the table in descending order of frequency Generate a table of counts, frequences and missing observations
fre
varlist
if
in
weight
, options
nomissing omit missing values from the table nolabel omit labels include(numlist) include only values specied in numlist ascending display rows in ascending order of frequency descending display rows in discending order of frequency
C.I.D.E.
40 / 59
January 11-15, 2010
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Discrete Variables
Cross-tabulation of 2 variables
tabulate varname1 varname2

chi2 report Pearsons exact (#)
2
if
in
weight
, options
report Fishers exact test
gamma report Goodman and Kruskals gamma column report the relative frequency within its column of each cell row report the relative frequency within its row of each cell cell report the relative frequency of each cell nofreq do not display frequencies (use only with column, row or cell) summarize(varname3) report summary statistics (mean, sd) for varname3 Cross-tabulation of more than 2 variables, by values
tab2 varlist
if
in
weight
, options
C.I.D.E.
January 11-15, 2010 41 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Tables of Statistics
table
table rowvar colvar supercolvar if in weight , options
where in rowvar colvar supercolvar
we put categorical variables (up to 3).
by(superrowvarlist) variables to be treated as superrows (up to 4) contents(clist) contents of the tables cells, where clist may contain up to 5 statistics mean varname mean sd varname standard deviation sum varname sum n varname count of nonmissing observations max, min varname maximum and minimum value median varname median p1... p99 varname percentiles iqr varname interquartile range (p75-p25)
C.I.D.E.
January 11-15, 2010 42 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Tables of Statistics
tabstat
tabstat varlist if in weight , by(varname) options
where in varlist we place a list of continuous variables, in by(varname) a categorical variable and among the options in statistics() we can choose: mean n sum max, min sd cv coecient of variation (sd/mean) semean standard error of mean (sd/sqrt(n)) skewness index of skewness kurtosis index of kurtosis p1... p99 range = max - min iqr interquartile range = p75 - p25
C.I.D.E.
January 11-15, 2010 43 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Sample Tests
Tests apply to variables and allow to compare statistical signicance of estimates against a null hp on the whole sample observed or for its subgroups. You can also use the command to perform tests under subgroups mean equality for the same variable, once a dicotomus variable identifying groups is selected (use missing values in this dummy varible to select only two subgroups which do not exhaust the sample dimension n). You can use the test_command i to obtain t-tests from inputted data (n, sd, means) that you like to compare. Other more specic tests can be downloaded and installed. Here is a list of general features: Test for means equality: help ttest Performs t-test for 1) one variable sample mean equality to a constant 2) one variable two subsample means equality 3) two variables means equality 4) two variables two subsample means equality. You can specify distributions. Test for standard deviations equality: help sdtest Performs t-test for 1) one variable sample sd equality to a constant 2) one variable two subsample sd equality 3) two variables sd equality 4) two variables two subsample sd equality. You can specify distributions.
January 11-15, 2010
C.I.D.E.
44 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Graphs
help graph This is the general command for all graphs. From here you can start searching the graphic style that you need, for univariate or multivariate representations, as well as statistical constructions. This is the broadest family of graphics. help twoway This command applies to bivariate graphics. The objective is to obtain a classical 2 coordinates graph, like scatter plots, connected points, condence intervals, regression t, distributions... One graph may contain dierent series (ex: time against income and consumption) or dierent objects (ex: observed and predicted values). Once twoway is declared, you have to select (in pairs) the variables that you want to put in the same graph and the type of graph linking the two variables (ex: scatter; connected; lfit; tsline; bar; spike; mband; lpoly; function;...). In this way, graphs are built sequentially and all the objects will appear in the same space. You can add graphs options by looking at the help. Using by(), you obtain separeted graphs according to the variable you want to be conditioned to. Options and in, if must be specied for each graphic tool you use, bacause they refer to a particular set of data in use. C.I.D.E.
January 11-15, 2010 45 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Regression Analysis
Stata econometrics models can be included into 5 large families, but only the rst one will be analysed. You are invited to read on the help the full characterization of commands. In each help le you also nd examples and interpretations of output results. 1) Cross-section econometrics You are invited to look at help regress for the admissible regression commands, including OLS, IV, limited dependent variables methods, treatment eects models, censoring and selection bias corrections, 3SLS, systems of equations, quantile regression. Using the command help regress_postestimation you also obtain information on post-estimation tests and model application syntax. 2) Time series econometrics You are invited to look at help time, you will nd all the list of commands associated with time series estimations, and how to build econometrics models in Stata. In particular, help tsset can be used to declare a time series structure of your data, and than proceed with usual regression techniques. Using the command help regress_postestimationts you also obtain information on post-estimation tests and model application syntax. C.I.D.E.
January 11-15, 2010 46 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Regression Analysis
3) Panel data econometrics You are invited to look at help xt, you will nd all the list of commands associated with the panel dimesion of a database. In particular, help xtreg oers a wide explanation of panel-data analysis techniques, while with help xtreg_postestimation you also obtain information on post-estimation tests and model application syntax. 4) Survey data analysis See help survey for all the details on data setting and regression techniques 5) Spatial econometrics See help spatreg or spatwmat for geographically located data. Commands available for Stata 10 or superior. C.I.D.E.
January 11-15, 2010 47 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Regression Analysis
Common models: OLS, IV, probit, multiple logit
regress depvar if in weight indepvars if in weight , noc options
ivregress estimator depvar probit depvar mlogit depvar indepvars indepvars
varlist1 if if in in
(varlist2=varlist_iv) weight weight , options , options
, noc options
options refers to regression specic options or estimation correction (robust se, constant...) estimator IV can be performed by 2SLS, GMM or limited info max likelihood.
C.I.D.E.
January 11-15, 2010 48 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Regression Anlysis
Other regression models: List I areg an easier way to t regressions with many dummy variables arch regression models with ARCH errors arima ARIMA models boxcox Box-Cox regression models cnreg censored-normal regression cnsreg constrained linear regression eivreg errors-in-variables regression frontier stochastic frontier models heckman Heckman selection model intreg interval regression ivregress single-equation instrumental-variables regression ivtobit tobit regression with endogenous variables newey regression with Newey-West standard errors qreg quantile (including median) regression
C.I.D.E.
January 11-15, 2010 49 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Regression Anlysis
Other regression models: List II tobit tobit regression treatreg treatment-eects model truncreg truncated regression xtabond Arellano-Bond linear dynamic panel-data estimation xtdpd linear dynamic panel-data estimation xtfrontier panel-data stochastic frontier model xtgls panel-data GLS models xthtaylor Hausman-Taylor estimator for error-components models xtintreg panel-data interval regression models xtivreg panel-data instrumental variables (2SLS) regression xtpcse linear regression with panel-corrected standard errors xtreg xed- and random-eects linear models xtregar xed- and random-eects linear models with an AR(1) disturbance xttobit panel-data tobit models
January 11-15, 2010
C.I.D.E.
50 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Post Estimation
Any Post Estimation command must be used immediately after a regression model, and in any case it refers to last estimates stored in Stata memory. You can save any model with a name and then proceed in post estimations recalling the model name. Predict calculates predictions, residuals, inuence statistics, and the like after estimation Predict regression output as new data
predict
statistic xb linear prediction; the default residuals residuals rstandard standardized residuals rstudent studentized (jackknifed) residuals stdp standard error of the linear prediction stdr standard error of the residual
type
newvar
if
in
, statistic
C.I.D.E.
January 11-15, 2010 51 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Post Estimation
Test linear hypothesis on coecients test coeist test exp=exp =... Test on residuals: normality and zero-mean sktest varname ttest varname = 0 acatter/mband Test on residuals: heteroskedasticity by graph and tests rvplot estat hettest estat imtest whitetst C.I.D.E.
January 11-15, 2010 52 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Post Estimation
Test on variables: inuential values
dfbeta
Logic of the dfbeta test:

1 2 3
Estimation with the complete sample Estimation without the i-th observation Comparison
Test on variables: collinearity

vif collin
NOTE: values of VIF > 10 deserve attention

C.I.D.E.
January 11-15, 2010 53 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Output: Interpreting
Results from computations are showed in the Results Window and stored in memory by the .log le. You can paste your output directly on your paper (or in other spreadsheet to rework them) by using the tables reported in the .log le. Stata oers other opportunites to list estimates and coecients in a proper way (ex. with standard errors between brackets) whithout requiring additional changes in estimates values. In this Programming section we will see how output coecients, vectors (ex. the regression parameters), matrices (ex. the variance-covariance matrix of regression coecients) or scalars (ex. the mean of a variable, its standard deviation, or an R2 estimation) can be properly represented and used. By matrix algebra, we can perform advanced econometrics on data. Remember that all the syntax previously presented applies exclusively to data stored in memory. To apply this syntax to your output coecients (for example compute the average of average values obtained by bootstrapping) you need to transform your results in data format (i.e. add columns). C.I.D.E.
January 11-15, 2010 54 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Show Estimates
Your code must be operable even after that changes in your data occurs (by merge or append). Therefore, you need to perform your tests or report coecients independently from your output windows results (ex: t-test for means). Stata allows to save the estimates as scalars or vectors/matrices. The general procedure is the following: 1 Perform your statistical procedures or your estimations (like summarize or regress); 2 Save your results and estimates in the memory by using r() or e() respectively. In this way you generate a new object; 3 Use your objects or display them in the output les. Alternatively, save regression results
outreg
varlist
using lename
, options
text opt.: nol; title(); addn() coecients opt.: bdec(); coefstar signicance opt.: se | p | ci | beta; bracket; 3aster stat opt.: addstat(r2, N, F); xstats other opt.: replace; append
C.I.D.E.
January 11-15, 2010 55 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Show Estimates
See help postest for a general description of post estimations commands. help estimates provides all the info to store, use and report estimates. All commands work i there is some result stored in the memory. Show estimates After statistics: see help return After regressions: see help ereturn Scalar dene, after summary statistics or table are reported
scalar define scalar_name = exp
scalar list scalar_list scalar drop scalar_list C.I.D.E.

January 11-15, 2010 56 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Show Estimates
After a regression command, use: To store estimates in the memory (use name of your model)
estimates store name estimates dir scalar drop name_list
, nocopy
To report statistics and estimates tables from model name
estimates stats namelist estimates table namelist
, n(#) , options
stats(N r2 ll chi2 aic bic rank) reports statistics in the table; keep(coeflist); drop(coeflist) to eliminate some coe from the table; b(%9.2f), se, t, p specify the format of coe reported and additioal stats;
To use results
t(names); r(coef); r(stats)
C.I.D.E.
January 11-15, 2010 57 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Work with Data as Matrices

Data can be exported as a matrix n k (a vector is a single column matrix), any statistics or estimation can be performed and result listed in the matrix format. New data created (ex., predictions) can be transformed in data from matrices. If you operate with matrices, use algebra to compute statistics (ex. X = (e e)1 e X), any other command will not be working. If you transform matrices in data (see the data editor), you can use previously seen commands. Use help matrix for an exhaustive review of commands to be used. To perform more complicated computation or build your program, use the Mata environment which allows to use Stata more interactively. See help mata for review of commands, available in Stata 9 or superior.
C.I.D.E.
January 11-15, 2010 58 / 59
Introduction
-1- Environment
-2.a- System Syntax
-3- Output
Work with Data as Matrices

Set the size of a matrix
set matsize #
Data Matrix; Matrix Data; Input Matrix

mkmat varlist if in , matrix(matname) nomissing
rownames(varname) svmat matname , names( col | eqcol | matcol | string ) matrix input matname = (#[, #...][\#[, #...][\[...]]])
Operations with matrices

matrix define matname = matrix_expression C.I.D.E.
January 11-15, 2010 59 / 59

Stata Tutorial

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Stata Tutorial

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Stata Tutorial

Uploaded by

Copyright:

Available Formats

Introduction

-2.a- System Syntax

-2.b- Statistical Syntax

Stata Tutorial IT2010

January 11-15, 2010 Alba di Canazei

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Organization of the Directory

* You are here! *

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

and unocial ones: http://www.ats.ucla.edu/stat/stata/ http://www.nyu.edu/its/statistics/Docs/Intro_stata5.pdf http://www.eui.eu/Personal/Researchers/decio/PS/Stata.pdf http://leuven.economists.nl/stata/stataintro.pdf

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Organization of the Tutorial

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Organization of the Tutorial

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Setting the Environment

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Setting the Environment

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Files Extentions in Stata

-2.a- System Syntax

-2.b- Statistical Syntax

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial

-2.a- System Syntax

-2.b- Statistical Syntax

Prepare Your .do File

Francesco Andreoli - Andrea Bonfatti (Universit di Verona) Stata Tutorial