Creating publication quality tables in Stata (2023)

Stata tables are generally clear and informative. However, they do not have the format or aesthetic quality typically used in publications. Several Stata users have written programs that produce publication-quality tables. This article will discussestablish(think "estimate table") by Ben Jann. OestablishThe command takes the results of previous estimates or other commands, puts them into a release quality table, and then saves that table in a format you can use directly in your work, such as . B. RTF or LaTeX. One of the main topics of this article is creatingRegression Result Tables,summary statistics tables, zfrequency tables.

The package is

Öestablishcommand is just one member of a family of commands, or packages, calledI am🇧🇷 Actually,establishis just a "wrapper" for an invoked commandI am. ÖI amcommand gives you full control over the table to be created, but requires complexity and flexibilityI amit is quite difficult to use. Oestablishexecuted commandsI amfor you and regulates many detailsI amrequires, so you can create the most common spreadsheets with relative ease. we will also discussestpost, which inserts results into a form as summary statisticsestablishcan work with. The ability to process summary statistics and frequencies in addition to regression results is one of the reasons we decided to focus on this articleestablish.

In the table creation workflow

Remember that you always have an alternative to use.establish: Simply create the tables you want in Word or your favorite word processor by copying and pasting the required numbers from the Stata output. This is time consuming and troublesome. On the other hand, you're trying to figure out how to get thereestablishProviding the desired table can also be time consuming and there is no guarantee that the exact table desired can be manufactured. Be sure to consider the possibility that manually creating a given table might be faster than using itestablish🇧🇷 A lot depends on how many tables you need to create and how many numbers they contain. If you canestablishto give you something close to what you want but spend too much time figuring out how to get itExactlywhat you want, just consider editing what you have.

Most people will find it easier to first come up with a set of (hopefully) end results and then work out how to present them. Running is not recommendedestablishuntil you're reasonably sure you've got the results you want to publish.

install rod

SinceI amThe package is not part of the official Stata, you need to install it before you can use it. It is available from the Statistical Software Components (SSC) archive and can be installed withssc-InstallationNo Stata command:

ssc-Installationsstatus

You only have to do this once - don't include this command in your search files.

Check for updates regularly by usingTo update.

basics

ÖestablishThe command needs some results to be able to act on it. So load the automatic data set that comes with Stata and run a simple regression:

sysuse auto
foreign reg mpg weight

You can see the basic function of esttab just by running it with no options:

establish

----------------------- (1) mpg ------------------- ---- - -----Weight -0.00659*** (-10.34) International -1.650 (-1.53) _cons 41.68*** (19.25) --------- - ----- ---- ---------N 74 ----------- t statistic in parentheses *p<0.05, **p<0.01, ***p<0.001

This places the model results in a table in the Stata Results window. The preview in the results window is useful for testing a table specification, but once you have what you want, you must do itestablishSave it in the file format you use for your paper. The standard table contains many of the features you would expect from a table of regression results in a newspaper article, including rounded coefficients and asterisks for significance. Note, however, that the numbers in parentheses are t-statistics. Use thecomOption if you want to replace them with standard errors:

found, s

----------------------- (1) mpg ------------------- ---- - ----- Weight -0.00659*** (0.000637) International -1.650 (1.076) _cons 41.68*** (2.166) --------------- - - ---- -------N 74 ----------------------------Standard error in brackets* p<0 .05, **p<0.01, ***p<0.001

ÖestablishThe command uses the current contents of the e() vector (information about the last estimation command), not the results of the last regression displayed. When you run oneLogitcommand with theorThe Stata option shows the odds ratios:

alien logit mpg, or

Logistic Regression Number of Observations = 74 LR chi2(1) = 11.49 Probability > chi2 = 0.0007Log-Likelihood = -39.28864 Pseudo-R2 = 0.1276------------- - ------- - ------------------------------------------------------- -- ------- - ----- Aliens | Default Odds Ratio. err. zP>|z| [95% Conf Range]--------------------------------------+------------------------------ ----- ---------- ------------------------ mpg | 1.173232 .0616975 3.04 0.002 1.05833 1.300608 _cons | 0.0125396 0.0151891 -3.62 0.000 0.0011674 0.1346911 ------------------- ----- ----- ---- ------------------------------

However,e(b)still contains the coefficients and by default this is whatestablishIs displayed. It also flags the test stats as t-stats instead of z-stats like the statsLogitoutput does:

establish

----------------------- (1) Foreigners ------------------- ---- - -----foreign mpg 0.160** (3.04) _cons -4.379*** (-3.62) ---------------------- ---- - -N 74 ---------------------------- t-statistic in brackets* p<0.05, ** p<0.01 , ***p<0.001

If you want odds ratios in your spreadsheet, typeestablishathe picture(exponential form) option. If you want the table to show "z stats in parentheses" instead of "t", use thezoption (note that thezoption does not change the numbers in any way):

found, form e.g

----------------------- (1) Foreigners ------------------- ---- ------foreign mpg 1,173**(3.04)-----------------------N 74--------- -- - ----------------Potential coefficients; z-Statistics in parentheses *p<0.05, **p<0.01, ***p<0.001

Specifying thethe picturePrompts optionestablishRemove the constant term from the table because it doesn't make much sense to talk about the odds ratio of the constants. However, you can override this behavior by specifying theConstantlyPossibility.

Save the table in your paper format

To save a table as a Rich Text Format (RTF) file, addusefilename.rtfto the command, just before the comma for options. Add those toosubstituteOption to allow previous versions of the file to be overwritten.

build with logit.rtf, replace eform z

The rich text format contains formatting information as well as the text itself and can be opened directly by Word and other word processing programs.Click here to see what the RTF file looks like.

The process of saving the spreadsheet as a LaTeX file is the same: just replace.rtfcom.tex🇧🇷 There are some special options that apply to LaTeX, such asFragmentto create a table fragment that can be added to an existing table. HTML (.html) is another useful format option, and there are many more.

You can save the spreadsheet as a comma-separated variables (CSV) file, which can be easily read in Excel by typing thefile extension for.csv🇧🇷 However, think carefully about whether what you intend to do in Excel can't be done better (and above all reproducibly) in Stata.

Tables with several models

To create a table that contains the estimates of multiple models, the first step is to run each model and save its estimates for future use. You can save the estimates using the official Stata commandappraisal shop, mostly abbreviatedthat's it, or with the variantThat's itincluded in itI amPackage. ISLANDThat's itThe variant adds some functionality, but we won't be using any of them in this article, so it doesn't matter which command you use. The basic syntax is the same: the command, then the name you want to give this set of estimates. Use this to create a set of nested models:

foreign registration mpg
it is sto m1
reg mpg foreign weight
it is only m2
reg mpg foreign weight offset gear_ratio
this is m3

TerestablishTo create a table based on a single set of stored estimates, just specify the name of the estimates you want to use:

set up m1

But you are not limited to one set:

m1 m2 m3 festlegen

-------------------------------------------------- ---------- (1) (2) (3) mpg mpg mpg ---------------------- ------ ---------------------Foreigners 4,946*** -1,650 -2,246 (3.63) (-1.53) (-1.81) Weight - 0 .00659* ** -0.00675*** (-10.34) (-5.80) Displacement 0.00825 (0.72) Gear Ratio 2.058 (1.17) _cons 19.83*** 41, 68 *** 34.52*** (26.70) (19.25) (5.17) - --------------------- ---- ---------------- ---- ---------N 74 74 74 ------------- ---- ----------------- --- ------------------- t-Statistics in parentheses* p<0.05 , **p<0.01, ***p<0.001

Summary statistics (model level)

N (number of observations) for each model is shown by default, but you can add other model-level statistics. Options include R Square (r2), AIC (aic) and BIC (bik🇧🇷 All other scalars in the e() array can also be addedrise()Possibility. For example, you can add the F-statistic of the model, saved ase(F), with the optionScale (F)🇧🇷 You cannot control the order in which they are listed, but you can move N to the end withobslast🇧🇷 You can remove N completely withBeginner.

esttab m1 m2 m3, see aic obslast escalar(F) bic r2

-------------------------------------------------- ---------- (1) (2) (3) mpg mpg mpg ---------------------- ------ ---------------------Foreigners 4,946*** -1,650 -2,246 (1,362) (1,076) (1,240) Weight -0.00659* ** -0, 00675*** (0.000637) (0.00116) Displacement 0.00825 (0.0114) Gear Ratio 2.058 (1.755) _cons 19.83*** 41.68*** 34.52* ** (0.743) (2,166) (6,675)- ------ ------------------------------------------ ------ --- --- ---R Square 0.155 0.663 0.669 AIC 460.3 394.4 396.9 BIC 465.0 401.3 408.4 F 13, 18 69.75 34.94 N 74 74 74 ---- --------------- ---- ----------- -------------------- -Standard error in brackets * p<0.05, ** p< 0.01, ***p<0.001

Cell Statistics (Variable Level)

In addition to t-statistics, z-statistics and standard errorsestablishYou can use parentheses to enclose p-values ​​and confidence intervalspecioptions. You cannot have a smaller amount in parentheses with theNot(ohne t) Option.

You can also substitute main numbers. OBetaOption replaces them with standardized beta coefficients. Oa director()With the option you can replace them with any other quantity of thee()Vector.

If you prefer to have the statistic in parentheses on the same line as the coefficient, use theLargoPossibility.

Estab m1 m2 m3, amplo ci noobs

-------------------------------------------------- -------------------------------------------------- ------------------------ (1) (2) (3) mpg mpg mpg ---------- ---- ---------------------------------------------- ---- ---------------------------------------------- ---- ----------------International 4,946*** [2,230.7,661] -1,650 [-3,796.0,495] -2,246 [-4,719.0,227]Weight -0.00659* ** [-0.00786,-0.00532]-0.00675*** [-0.00907,-0.00443]shift 0.00825 [-0.0145,0.0310]gear ratio 2.058 [-1.444,5.559 ]_cons 19.83*** [18,35,21,68] 18,35,2 ,46,00] 34.52*** [21,21,47,84]-------- ---- ------------------------ -- ------ ----- ------------------------ - ------ ------ ----------------- 95% confidence intervals in brackets * p < 0.05, ** p < 0.01, *** p < 0.001

Titles, notes and labels

With you can give the table a general titleTitle()Possibility. Enter the desired title in brackets.

If you want to remove the note below explaining the numbers in brackets and the meaning of the stars, use theRemarksPossibility. If you want to add notes, use theadd notes ()Option with the desired notes in brackets. If you want multiple lines of notes, enclose each line in quotation marks.

By default, each model is identified in a table with a number and a title. If you don't want the number to appear, use theno numberPossibility. The template title is the template dependent variable name by default, but you can change the template title withTitle()🇧🇷 Each title is enclosed in parentheses with quotation marks, and the order must match the order in which the saved estimates are listed in the main command.

ÖLabelsoption saysestablishUse variable labels instead of variable names. This means you can control exactly how a variable is listed by changing its label - just make sure the label describes the variable appropriately, but not too long. The following labels illustrate some of the potential issues.

esttab m1 m2 m3, label without number title("MPG models") mtitle("model 1" "model 2" "model 3")

MPG Templates ------------------------------------------------ -------------- ------------ - -------------------- Model 1 model 2 Model 3 ----------- ------------ ------------------------ -------------- ------Vehicle Type 4,946** * -1.650 -2.246 (1.362) (1.076) (1.240) Weight (lbs.) -0.00659*** -0.00675*** (0.000637) (0.00116) Displacement ... in .) 0.00825 (0.0114) Gear Ratio 2.058 (1.755) ) Constant 19.83*** 41.68** * 34.52*** (0.743) (2.166) (6.675) ------- ------ -------------- ------- ----------------------- ------ -----Remarks 74 74 74 ----- ------- ----------------------- ------ -------------- ------- ------Standard error in brackets * p<0.05, ** p<0.01, *** p<0.001

If you don't want to change the actual variable labels, you can replace them with thecoeflabel()Possibility. Enclose in parentheses the variable name/label pairs you want to use. Any variable that you don't specify a label for is listed with its correct name.

esttab m1 m2 m3, coeflabel(foreign „Foreign Car“ deslocamento „Deslocamento“ gear_ratio „Gear Ratio“ _cons „Constant“)

-------------------------------------------------- ---------- (1) (2) (3) mpg mpg mpg ---------------------- ------ ---------------------Foreign car 4,946*** -1,650 -2,246 (3.63) (-1.53) (-1.81) Weight - 0.00659*** -0.00675*** (-10.34) (-5.80) displacement 0.00825 (0.72) gear ratio 2.058 (1.17) constant 19.83** * 41, 68*** 34.52*** (26.70) (19.25) (5.17) - ------------------------ ------------------- ------- --------N 74 74 74 ------------ -------------- ------------ ------------ t-statistics in parentheses* p<0.05, * *p<0.01, ***p<0.001

Formate

In general, you can change the format of a number by putting your desired format in parentheses after the option that prompts you to display that number. Useb()to format the betas andt()format statistics t.

estab m1 m2 m3, b(%9.1f) t(%9.1f) r2(%9.6f)

-------------------------------------------------- ---------- (1) (2) (3) mpg mpg mpg ---------------------- ------ ---------------------Foreigners 4.9*** -1.7 -2.2 (3.6) (-1, 5) (- 1 ,8) Weight - 0.0066*** -0.0068*** (-10) (-5.8) Displacement 0.0082 (0.72) Gear ratio 2.1 (1.2) ) _cons 20* ** 42*** 35*** (27 ) (19) (5,2) ----------------------- ------ ------------N 74 74 74 R square 0.154762 0.662703 0.669463 - ------- ------------ ----------- ------ ------- -- t-statistics in parentheses * p<0.05, ** * p<0.01, *** p<0.001

stars and meaning

ÖStern()This option allows you to control when stars are used. Inside the brackets you put a list of characters together with the numeric threshold below which you apply a coefficient. The pattern is equivalent to:

Stern(* 0,05 ** 0,01 *** 0,001)

Notice thatStern()Pay attention to the numbers and how you format them: if you omit the leading zeros, they won't appear in the table.

estab m1 m2 m3, p Stern(+0,1 * 0,05 ** 0,01)

-------------------------------------------------- ------- (1) (2) (3) mpg mpg mpg ------------------------------- ------ ---------------------Foreigners 4,946** -1,650 -2,246+ (0.001) (0.130) (0.074) Weight -0.00659 ** - 0.00675** (0.000) (0.000) Displacement 0.00825 (0.472) Gear Ratio 2.058 (0.245) _cons 19.83** 41.68** 34.52** (0.000) (0.000) (0.000 ) ------ ------ --------------------------------- ---- -N 74 74 74 - ---------------------------- ------------ ---- --------p-values ​​in brackets+ p<0.1, * p<0.05, ** p<0.01

Summary statistics tables

ÖestablishThe command is used to extract information from the e() array, which is only used by estimation commands. However,estposttakes the results of the r() array used by other commands and sends them to the e() array. This allowsestablishto create tables based on these results, but generally you need to provide more guidance on what that table should contain.

To store the results of a command in e(), set theestpostcommand before:

estpost total foreign price mpg

The resulting table shall give the official name of each set. You will use these names belowestablishcommands.

🇺🇸 e(count) e(sum_w) e(average) e(Var) e(sd) e(min) e(max) e(sum) -------------+-- -- ------------------------------------------------ ------------------ ------------------ Preis | 74 74 6165.257 8699526 2949.496 3291 15906 456229 Ausland | 74 74 .2972973 .2117734 .4601885 0 1 22 mpg | 74 74 21.2973 33.47205 5.785503 12 41 1576

When working with regression resultsestablishknow thate(b)is the principal interest amount and constructs the table accordingly. With summary statistics you have to sayestablishwhat the table should containCell()Possibility. This is technically an option forI amrather thanestablish, Then whatestablishwill pass onI amwhile still doing some of the other work for you. However, if you want to read the full documentation for theCell()option that you must enterhelp amrather thanhelp with construction.

If you want a fair table of averages, usecell (average):

stitch, cell (average)

--------------------------------------- (1) means--------------------- ----Price 6165.257Abroad .2972973mpg 21.2973----------------------N 74--------- --- --- ----------

You can specify multiple quantities:

esttab, cell (media SD)

------------------------- (1) Average/SD --------------- ---- --Price 6165.257 2949.496Abroad .2972973 .4601885mpg 21.2973 5.785503------------------------N 74------------ -------------

If you want the sets to appear on a single line, you can group them with quotes or parentheses. The following commands are equivalent:

esttab, cell ("media SD")
esttab, cell ((media SD))

-------------------------------------- (1) Average SD------- ------------------------Price 6165.257 2949.496Abroad .2972973 .4601885mpg 21.2973 5.785503-------- ------- ----------------------N 74 ------------------- ------- ------------

Note that the quotes do not indicate strings in this case!

Model numbers and model titles make little sense for this table (especially since the title is blank at this point), so consider removing them as wellno numbereName:

lose, cell((mean sd)) nonumber titlename

-------------------------------------- means sd---------- -----------------------Price 6165.257 2949.496Abroad .2972973 .4601885mpg 21.2973 5.785503------------ ----- ---------------------N 74 ---------------------- ----- -----------

To control the numeric format of the results listed inCell()benutze ofmt()Possibility:

esttab , cell ( ( average ( fmt ( %9.1f )) sd ( fmt ( %9.2f )))) nonumber nometítulo

-------------------------------------- means sd---------- ----------------------- Price 6165.3 2949.50 International 0.3 0.46 mpg 21.3 5.79 ------ -------- ---- ---------------------N 74 --------------- --------- --- ------------

There are many other options. A useful addition to this table isParfor brackets:

create, cell((means sd(pair))) nonumber titlename

-------------------------------------- means sd---------- -----------------------Price 6165.257 (2949.496)International .2972973 (.4601885)MPG 21.2973 (5.785503)----- ----- ----------------------------N 74 --------------- ---- --------------------

The labeling of the column headings also leaves a lot to be desired. You can replace them with alabels()Option associated with each quantity inCell()🇧🇷 This is different than generalLabelsoption that saysestablishto replace the variable names at the beginning of each line with the variable labels. You are welcome to use both (orcoeflabel()to define the row labels yourself):

esttab, cell((mean(label(Mean)) sd(par label(standard deviation)))) label nonumber nomtitle

---------------------------------------------- Standard average D~ n ------------------------------------ Price 6165.257 ( 2949.496) Vehicle type 0.2972973 (0.4601885) Mileage (mpg) 21.2973 (5.785503)------------------------------------- -------Notes 74 --------------------------------- --- ----- ----

The problem now is that the "standard deviation" had to be truncated because its column is not wide enough. With you can set the column widthModellbreite ()option (remember that when dealing with regression results, each column is a model). If you enclose a single number in parentheses, the character width of all columns is set to that number. If you specify a list of numbers, they are applied to the columns in order:

esttab, modelwidth(10 20) cell((mean(label(Mean)) sd(par label(standard deviation)))) label nomtitle nonumber

-------------------------------------------------- -- Mean standard deviation --------------------------------------------- - ------Price 6165,257 (2949,496)Car Type 0.2972973 (0.4601885)Mileage (mpg) 21.2973 (5.785503)-------------- ---- ----------------------------------Notes 74 ----------- ---- ---------------------------------

Admittedly, in plain language, this will never achieve publication quality. but consider thatRTF-Version, created by:

verwendet mean.rtf, modelwidth(10 20) cell((mean(label(Mean)) sd(par label(Standard Deviation)))) label nomtitle nonumber replace

frequency tables

The creation of frequency tables is also usage dependent.estpostto write the results into the e() array:

estpost tab rep78 foreign

foreign | rep78 | e(b) e(pct) e(colpct) e(rowpct) -------------+-------------------- ------------------------Domestic | 1 | 2 2.898551 4.166667 100 2 | 8 11.5942 16.66667 100 3 | 27 39.13043 56.25 90 4 | 9 13.04348 18.75 50 5 | 2 2.898551 4.166667 18.18182 Total | 48 69.56522 100 69.56522 ---------------------------+------------------------------- - -----------Aliens | 1 | 0 0 0 0 2 | 0 0 0 0 3 | 3 4.347826 14.28571 10 4 | 9 13.04348 42.85714 50 5 | 9 13.04348 42.85714 81.81818 Total | 21 30.43478 100 30.43478 ---------------------------+------------------------------- - -----------Total | 1 | 2 2.898551 2.898551 100 2 | 8 11.5942 11.5942 100 3 | 30 43.47826 43.47826 100 4 | 18 26.08696 26.08696 100 5 | 11 15.94203 15.94203 100 Total | 69 100 100 100

These are the same numbers you would receiveabalone, just arranged differently. Note that the frequencies themselves are callede(b), but we will use it anywayCell()otherwiseestablishtreats them as regression coefficients:

stable, cell(b)

------------------------ (1) b-------------------- - ----Domestic 1 22 83 274 95 2Total 48--------------------Foreigners 1 02 03 34 95 9Total 21---- - ---- ----------------Total 1 22 83 304 185 11Total 69-------------------- - ---N 69-------------

Model number, blank model title, and column label (b) are all useless here, so remove the number and title and change the caption withcollaborations()🇧🇷 You can also completely remove the column labelcollaborations(none).

esttab, cell(b) nonumber nomtitle collabels(frequency)

------------------------- Frequency------------------------------------- -Domestic 1 22 83 274 95 2Total 48 -------------------- Abroad 1 02 03 34 95 9 Total 21 ------- - ---- -------------Total 1 22 83 304 185 11Total 69----------------------- - No. 69-- ----------------------

ÖdestackThe option converts the three sections into columns:

esttab, cell(b) unstack nonumber nomtitle collabels(nenhum)

-------------------------------------------------- - Total foreign domestic ---------------------------------------------- - ----1 2 0 22 8 0 83 27 3 304 9 9 185 2 9 11Total 48 21 69------------- ------------- - ------------N 69 ------------------------ ------------ - -----------

To control the labeling of the row variable, useEQLabels(), Then whatestablishThink of it as the left side of an equation (rememberestablishwas built for models). So you have to use thoselhs()sub option insideEQLabels()🇧🇷 You can customize the available space for the label withvarwidth():

esttab, cell(b) eqlabels(, lhs("Repair Record")) varwidth(15) Unstack nonumber nomtitle collabels(nenhum)

-------------------------------------------------- ----Record of total repairs domestic overseas ------------------------- -- ----1 2 0 22 8 0 83 27 3 304 9 9 185 2 9 11 Total 48 21 69 ------ --------- --------- -------------N 69 ------------- -------- -------------- --------------------

You can add additional quantitiesCell()and control its appearance and structure with all the tools we discussed in the summary statistics section. Consider adding a note to explain what each number represents with theUse()Possibility:

esttab, cell(b rowpct(fmt(%5.1f) par)) note(Row Percentages in Parentheses) unstack nonumber nomtitle collabels(none) eqlabels(, lhs("Repair Record")) varwidth(15)

-------------------------------------------------- ----Record of total repairs domestic overseas ------------------------- ----1 2 0 2 (100.0) (0.0) (100.0)2 8 0 8 (100.0) (0.0) (100.0)3 27 3 30 (90 .0 ) (10.0) (100.0)4 9 9 18 (50.0) (50.0) (100.0)5 2 9 11 (18.2) (81.8) (100.0 )Total 48 21 69 (69.6) (30.4) (100.0)-------- ---------- ------------ --------N69------------------------ -- -------------------------- - Percentages of rows in parentheses

That's just a fraction of whatestablish(much lessI am) can. To learn more, we recommend reading theArticle in Stata Journal in which it was featured🇧🇷 Enter for syntax detailshelp with constructionand orhelp am.

Last revision: 03/26/2015