{smcl}
{hline}
help for {hi:mktab}{right:Nick Winter}
{hline}
{title:Print table of estimates in delimited or screen-presentation format}
Basic syntax:
{p 5 11 2 }{cmd:mktab} {cmd:(}{it:depvar1 varlist1}{cmd:)}
{cmd:(}{it:depvar2 varlist2}{cmd:)}
{it:...} {cmd:(}{it:depvarN varlistN}{cmd:)} [{it:weight}]
[{cmd:if} {it:exp}] [{cmd:in} {it:range}] [{cmd:,} {it:options} ]
Full syntax:
{p 5 11 2}{cmd:mktab} {cmd:(}[{it:eqname1}{cmd::}] {it:depvar1a} [{it:depvar1b}
{it:...} {cmd:=}] {it:varlist1} [{cmd:,} {cmdab:noc:onstant}]{cmd:)}{p_end}
{p 11 11 2}{cmd:(}[{it:eqname2}{cmd::}] {it:depvar2a} [{it:depvar2b} {it:...}
{cmd:=}] {it:varlist2} [{cmd:,} {cmdab:noc:onstant}]{cmd:)}{p_end}
{p 11}{it:...}{p_end}
{p 11 11 2}{cmd:(}[{it:eqnameN}{cmd::}] {it:depvarNa} [{it:depvarNb} {it:...}
{cmd:=}] {it:varlistN} [{cmd:,} {cmdab:noc:onstant}]{cmd:)}{p_end}
{p 11 11 2}[{it:weight}] [{cmd:if} {it:exp}] [{cmd:in} {it:range}] [{cmd:,}
{it:options} ]
Options:
{p 11 11 2}{cmdab:p:rint(}{it:varlist}{cmd:)}
{cmdab:l:og(}{it:filename}[, {cmd:append} | {cmd:replace}]{cmd:)}
{cmd:cmd(}{it:cmdname}{cmd:)}
{cmdab:a:ux(}{it:name}[{cmd:=}{it:label}][{cmd:,}{it:name}[{cmd:=}{it:label}] {it:...} ]{cmd:)}
{cmdab:e:st(}{it:name}[{cmd:=}{it:label}][{cmd:,}{it:name}[{cmd:=}{it:label}] {it:...} ]{cmd:)}
{cmdab:f:lag(}{it:#}{cmd:=}{it:label}[{cmd:,}{it:#}{cmd:=}{it:label ...} ]{cmd:)}
{cmdab:t:ag(}{it:string}{cmd:)}
{cmdab:d:elimit(}{it:string}{cmd:)}
{cmdab:n:oisily}
{cmdab:c:ontinue}
{cmdab:xl:abel} {cmdab:yl:abel}
{cmdab:t1:title} {cmdab:t2:title} {cmdab:b1:title} {cmdab:b2:title}
{cmdab:notit:le} {cmdab:nobtit:le}
{cmd:mif(}{it:conditions}{cmd:)}
{cmdab:mifl:abel(}{it:labels}{cmd:)}
{cmdab:pl:evels}
{cmdab:nose}
{cmdab:bf:mt(%}{it:fmt}{cmd:)}
{cmdab:sf:mt(%}{it:fmt}{cmd:)}
{cmdab:ef:mt(%}{it:fmt}{cmd:)}
{cmdab:pf:mt(%}{it:fmt}{cmd:)}
{cmdab:onet:ail}
{cmdab:sc:reen} {cmdab:vsp:ace(}{it:#}{cmd:)}
{cmdab:con:nect}
{cmd:notags}
{cmd:latex} {cmdab:nocap:tion}
{cmd:html}
{it:command_options} ]
{title:Description}
{p}{cmd:mktab} estimates one or more single-equation models, and
prints the results in tab- (or other) delimited format for importation into a
spreadsheet or word-processor. This facilitates creating tables in
"journal-article" format, i.e. with standard errors in parentheses below
parameter estimates, significance flagged, and so on.
{p}{cmd:mktab} estimates each model specified. It then produces a delimited
table with the dependent variables across the top, the RHS variables (and
auxiliary parameter and returned estimates results) down the left, and
the coefficients and standard errors in the cells. Each column corresponds
to a single model; each row to a single RHS variable.
{p}{cmd:mktab} is similar to John Gallup's exemplary {help outreg}. The primary differences
are that it estimates all the models and generates the table with a single command,
and that it does not require preserving and restoring the data. In addition, it
has an option to display output in space-delimited columns on the screen.
{cmd:if}, {cmd:in} and {help weights} are supported.
{p}This command has been tested with {help regress}, {help probit}, {help logit},
{help oprobit}, and {help ologit}, along with their {help svy:survey} counterparts.
It should work with any single-equation
estimator. Time series operators are not supported.
{title:Specification of Models to Run}
{p}Models may be specified with the "new" Stata 6.0 multi-equation syntax.
Thus, they can be entered individually, each surrounded by parentheses.
If several models involve the same RHS variables, they can be specified
together, with an equal sign separating the LHS and RHS variables, as in
{bind:{cmd:(mpg price weight = rep78 foreign)}.}
{title:Options}
{p 0 5 2}{cmd:print(}{it:varlist}{cmd:)} specifies the list (and ordering) of
RHS variables to include in the output. The default is to include all variables
from all models. Wildcards can be used to refer to variables in this list (and
will expand only to include variables in the models).
{p 0 5 2}{cmd:log(}{it:filename}{cmd:)} is the name of a log file for the results. The usual log
options, {cmd:replace} and {cmd:append} are valid. If a log file is open and
this option is specified, then the open log file will remain open for
any output other than the comma-delimited table (see {cmd:noisily}, below)
and will be re-opened after the table printed.
{p 0 5 2}{cmd:cmd(}{it:cmdname}{cmd:)} is the name of the command to run the model. The default
is {help regress}.
{p 0 5 2}{cmd:aux(}{it:string}{cmd:)} is a comma-separated list of auxillary estimates to print
immediately after varlist and in the listed order, e.g. {cmd:aux(_cons)},
optionally with labels, e.g. {cmd:aux(_cut1=Cut One,_cut2=Cut Two)}. They
need not exist for all the estimated models.
{p 0 5 2}{cmd:est(}{it:string}{cmd:)} is a comma-separated list of returned estimates to print
with optional labels (e.g. {cmd:est(N,r2=R Squared)}).
{p 0 5 2}{cmd:mif(}{it:conditions}{cmd:)} specifies a set of {help if}
conditions that vary across the
multiple equations. They are combined with the {help if} condition, if any,
with an &. Conditions should take the form:
{p 10 10 2}{cmd:mif(} [{it:stub} {cmd::} ] {it:cond1} [{cmd:\} {it:cond2} [{it:...}]]{cmd:)}
{p 5 5 2}where {it:stub} is pre-pended to each of the cond's. So, for example,
{bind:{cmd:mif(rep78== : 1 \ 2 \ 3 \ 4 \ 5 )}} would apply {bind:{cmd:if rep78==1}} to the
first equation estimates, {bind:{cmd:if rep78==2}} to the second, and so on.
The number of {cmd:mif()} conditions specified must match the number of
equations. However, if there is exactly one equation specified, and multiple
mif() conditions, then the single equation will be duplicated for each mif()
condition. This means that if you want to run the same model across multiple
subgroups, you need only specify the equation once, as in the example below.
{p 0 5 2}{cmd:miflabels(}{it:labels}{cmd:)} specifies a set of labels for the columns,
to replace those generated by default by the {cmd:mif()} option. For example
{p 10 10 2}{cmd:mktab (mpg = price weight length) , mif(foreign==1\foreign==0) miflabel("Foreign" "Domestic") }
{p 5 5 2}would label the two columns of output "Foreign" and "Domestic", rather than "foreign==1"
and "foreign==0"
{p 0 5 2}{cmd:flag(}{it:flaglist}{cmd:)} specifies a comma-separated
list of one or more significance (p)
levels, specified as integers, and corresponding symbols. The default is
{bind:{cmd:flag(1=**,5=*,10=^)}}, which labels p<0.01 with **; p<0.05 with *, and
p<0.10 with ^. Flag values must be in ascending order. To suppress
significance marking, specify {cmd:flag(0=*)}.
{p 0 5 2}{cmd:connect} indicates that the significance level flags (e.g. **) should be
connected with the coefficients, rather than separated into their own output columns.
{p 0 5 2}{cmd:tag(}{it:string}{cmd:)} is an identifying name that prints
in the third column of output
(e.g. {cmd:tag(Model One)}).
{p 0 5 2}{cmd:notags} indicates that line numbers, line types, and tag information should
not be included. This option is selected automatically when screen formatting
is requested.
{p 0 5 2}{cmd:delimit(}{it:string}{cmd:)} specifies a delimiter to be used to separate
the columns. The default is to tab-delimit the output.
{p 0 5 2}{cmd:noisily} indicates that the individual models should be displayed as they
are run. If a log file is open when {cmd:mktab} is executed, this
output will be sent to that log file; the delimited table will
be sent to the file indicated in the {cmd:log()} option.
{p 0 5 2}{cmd:continue} indicates that the master row numbers should continue from the
last invocation of {cmd:mktab}. This allows creation of a single log
file with multiple tables, one after the other.
{p 0 5 2}{cmd:xlabel} uses variable labels (if any) to label the RHS variables.
{p 0 5 2}{cmd:ylabel} uses variable labels (if any) to label the LHS variables.
{p 0 5 2}{cmdab:onet:itle} indicates that all the equations have the same LHS variable,
and that the column header (i.e., variable name or label) should span all the columns, rather
than being repeated for each.
{p 0 5 2}{cmd:t1title}, {cmd:t2title}, {cmd:b1title}, and {cmd:b2title} allow up to two lines of
titles at the top and bottom of the table.
{p 0 5 2}{cmd:notitle} and {cmd:nobtitle} suppress printing of the default top and bottom
title information. This is helpful in conjunction with the {cmd:continue()}
option.
{p 0 5 2}{cmd:plevels} indicates that p-levels should be displayed under the coefficients,
in place of standard errors.
{p 0 5 2}{cmd:nose} suppressed the display of standard errors (and p-levels)
{p 0 5 2}{cmd:bfmt} {cmd:sfmt} {cmd:efmt} and {cmd:pfmt} indicate the display format to use
for coefficients, standard errors, returned estimates, and p-values, respectively. The
defaults are %4.3f for coefficients and standard errors, %3.2f for estimates, and %5.4
for p-values.
{p 0 5 2}{cmd:onetail} indicates that one-tailed significance levels should be calculated.
The default is two-tailed.
{p 0 5 2}{cmd:screen} indicates that the table should be formatted for the screen, rather
than delimited. This will produce a space-delimited table that will display appropriately
in a fixed-pitch font. It is useful for reading model results during interactive use.
{p 0 5 2}{cmd:vspace} indicates the width of the column for variable names, when results
are formatted to the screen. The default is 20.
{p 0 5 2}{cmd:latex} indicates that the table should be formatted for inclusion in a LaTeX
file. See discussion of LaTeX, below.
{p 0 5 2}{cmd:nocaption} indicates that the table caption (taken from {cmd:t1title()})
should be omitted from the LaTeX output.
{p 0 5 2}{cmd:html} indicates that the table should be formatted as an HTML table.
{p 0 5 2}{it:command_options} can include any options appropriate to the command being run
(e.g. {help robust}).
{title:Output}
{cmd:mktab} outputs the following columns:
{p 0 18 2}{bind:1 row number: }consecutive numbering of the lines in the table as a whole,
which allow re-sorting into the correct order
{p 0 18 2}{bind:2 line type: }0 for header/title rows{p_end}
{p 18 18 2}1 for coefficients{p_end}
{p 18 18 2}2 for standard errors{p_end}
{p 18 18 2}3 for estimates, and{p_end}
{p 18 18 2}4 for footer rows.{p_end}
{p 18 18 2}These allow re-sorting of the data for formatting
{p 0 18 2}{bind:3 the tag: }Contents of the {cmd:tag()} option, to identify the table
{p 0 18 2}{bind:4 variable names: }The name of the variable for this row
(blank for std error rows)
{p 0 18 2}5+ Alternating columns of coefficients for each model, with significance
symbols for each coefficient.
Rows alternate between coefficients and their standard errors. Standard
errors are surrounded by parentheses.
{title:LaTeX output}
{p 0 0 2}The {cmd:latex} option specifies that the output table should be formatted as
a LaTeX table (i.e., in a tabular environment). The table is formatted with
coefficient and standard error lines decimal-aligned, and other lines centered.
Standard errors are printed in a smaller font under the coefficients.
{p 0 0 2}The resulting table makes use of two LaTeX packages: threewaytables and booktabs, so
be sure to include \usepackage{threeparttable,booktabs} in your LaTeX preamble.
{p 0 0 2}For example, the following command produces the following LaTeX output, and saves it in
the file mylog.tex:
{p 5 10 2}{cmd:. mktab (mpg price rep78 = gratio weight), log(mylog.tex) latex }{p_end}
{p 5 10 2}\begin{table}[ht]{p_end}
{p 5 10 2}\begin{center}{p_end}
{p 5 10 2}\begin{threeparttable}{p_end}
{p 5 10 2}\caption{{\em }}{p_end}
{p 5 10 2}\begin{tabular}{ l r@{}l r@{}l r@{}l }{p_end}
{p 5 10 2}{p_end}
{p 5 10 2}\toprule{p_end}
{p 5 10 2} & \multicolumn{2}{c}{{\em mpg}}& \multicolumn{2}{c}{{\em price}}& \multicolumn{2}{c}{{\em rep78}}{p_end}
{p 5 10 2}\\ {p_end}
{p 5 10 2} \cmidrule(lr){2-7}{p_end}
{p 5 10 2}{p_end}
{p 5 10 2}gear\_ratio&0.&099&1454.&284&0.&535{p_end}
{p 5 10 2}\\ {p_end}
{p 5 10 2} {p_end}
{p 5 10 2}&\raisebox{.7ex}[0pt]{\scriptsize (1.}&\raisebox{.7ex}[0pt]{\scriptsize 365)}{p_end}
{p 5 10 2}&\raisebox{.7ex}[0pt]{\scriptsize (978.}&\raisebox{.7ex}[0pt]{\scriptsize 092)}{p_end}
{p 5 10 2}&\raisebox{.7ex}[0pt]{\scriptsize (0.}&\raisebox{.7ex}[0pt]{\scriptsize 389)}{p_end}
{p 5 10 2}\\ {p_end}
{p 5 10 2}weight&--0.&006$^{**}$&2.&692$^{**}$&--0.&000{p_end}
{p 5 10 2}\\ {p_end}
{p 5 10 2} {p_end}
{p 5 10 2}&\raisebox{.7ex}[0pt]{\scriptsize (0.}&\raisebox{.7ex}[0pt]{\scriptsize 001)}{p_end}
{p 5 10 2}&\raisebox{.7ex}[0pt]{\scriptsize (0.}&\raisebox{.7ex}[0pt]{\scriptsize 574)}{p_end}
{p 5 10 2}&\raisebox{.7ex}[0pt]{\scriptsize (0.}&\raisebox{.7ex}[0pt]{\scriptsize 000)}{p_end}
{p 5 10 2}\\{p_end}
{p 5 10 2}\bottomrule{p_end}
{p 5 10 2}\end{tabular}{p_end}
{p 5 10 2}\begin{tablenotes}[flushleft]{p_end}
{p 5 10 2}\item \hspace{-0.2em}$^{**}$ p$<$0.01; $^{*}$ p$<$0.05; \raisebox{.7ex}[0pt]{\tiny $\wedge$} p$<$0.10 two tailed{p_end}
{p 5 10 2}{p_end}
{p 5 10 2}\end{tablenotes}{p_end}
{p 5 10 2}\end{threeparttable}{p_end}
{p 5 10 2}\end{center}{p_end}
{p 5 10 2}\end{table}{p_end}
{p 5 10 2}\clearpage{p_end}
{p 0 0 2}The following example LaTeX document would include this table:{p_end}
{p 5 10 2}\documentclass[8pt]{extarticle}{p_end}
{p 5 10 2} \usepackage{threeparttable,booktabs}{p_end}
{p 5 10 2} \oddsidemargin 0.0in{p_end}
{p 5 10 2} \evensidemargin 0.0in{p_end}
{p 5 10 2} \textwidth 6.5in{p_end}
{p 5 10 2} \topmargin 0.5in{p_end}
{p 5 10 2} \textheight 9.0in{p_end}
{p 5 10 2}\begin{document}{p_end}
{p 5 10 2} \input{mylog.tex}{p_end}
{p 5 10 2}\end{document}{p_end}
{p 0 0 2}In fact, this wrapper LaTeX file can be created automatically by mktab, if you issue the
following command:
{p 5 10 2}{cmd:. mktab wrapper , log(mylog.tex)}
{p 0 0 2}This creates a file _mylog.tex, which contains the wrapper listed above. This can be handy when
fine-tuning a table or set of tables.
{title:Examples}
{p 5 10 2}{cmd:. mktab (mpg price rep78 = gratio weight), log(mylog)}
{p 5 5 2}This runs three regressions (with DVs mpg, price and rep78), each
with RHS variables gratio and weight (and a constant). It prints the
coefficients (and their standard errors) into a file called mylog.log.
{p 5 10 2}{cmd:. mktab (vote90 vote92 = dem rep ideology) cmd(probit)}
{cmd:aux(_cons=Intercept) est(N,ll=Log Likelihood,r2_p=Pseudo R2)}
{cmd:tag(Table One) log(mylog, append) continue}
{p 5 5 2}This runs two probit models (DVs vote90 and vote92), each with three
dependent variables (dem, rep and ideology). The table includes
coefficients and standard errors for the three RHS variables and the
constant term (labeled Intercept), as well as estimates N, ll, and r2_p
(which are labeled "N", "LnL", and "Pseudo R2", respectively.)
It appends the table to mylog.log, and numbers the rows consecutively
with the previous table.
{p 5 10 2}{cmd:. mktab (rep78 mpg price) (rep78 mpg price weight), cmd(oprobit)}
{cmd:f(.1=**,1=*) pr(5) log(mylog, replace) xlab ylab delimit(,)}
{cmd:a(_cut1=Cut One,_cut2=Cut Two,_cut3=Cut Three,_cut4=Cut Four)}
{p 5 5 2}This runs two ordered probit models of rep78, one with independant
variables mpg and price, the other with mpg, price, and weight. It
prints the results with five decimal places, uses the variable labels
from the data set, flags coefficients p<0.001 with ** and
p<0.01 with *, and delimits the output with commas. This produces the following output:
{res} 1,0,,Estimates (using oprobit)
2,1,,Variable,Repair Record,,Repair Record,
3,1,, ,1978,,1978,
4,2,,Mileage (mpg),0.11342,**,0.05978,
5,3,,,(0.02897),,(0.04091),
6,2,,Price,0.00010, ,0.00014,
7,3,,,(0.00005),,(0.00005),
8,2,,Weight (lbs.),--, ,-0.00055,
9,3,,, ,,(0.00031),
10,2,,Cut One,0.96916, ,-1.66922,
11,3,, ,(0.82655),,(1.68018),
12,2,,Cut Two,1.83700, ,-0.78752,
13,3,, ,(0.79926),,(1.66034),
14,2,,Cut Three,3.22326,**,0.63996,
15,3,, ,(0.82574),,(1.65004),
16,2,,Cut Four,4.18628,**,1.65129,
17,3,, ,(0.87507),,(1.65440),
18,6,,** p<0.001, * p<0.01, two tailed
19,7,,
{text}
{p 5 10 2}{cmd:. mktab (rep78 mpg price) (rep78 mpg price weight), cmd(oprobit)}
{cmd:f(.1=**,1=*) pr(5) log(mylog, replace) xlab ylab screen}
{cmd:a(_cut1=Cut One,_cut2=Cut Two,_cut3=Cut Three,_cut4=Cut Four)}
{p 5 5 2}This runs the same models, but formats the output for the screen.
This produces the following output:
{text} Variable Repair Repair
Record Record
1978 1978
Mileage (mpg){res} 0.113{text}**{res} 0.060
(0.029) (0.041)
{text}Price{res} 0.000 0.000
(0.000) (0.000)
{text}Weight (lbs.){res} -- -0.000
(0.000)
{text}Cut One{res} 0.969 -1.669
(0.827) (1.680)
{text}Cut Two{res} 1.837 -0.788
(0.799) (1.660)
{text}Cut Three{res} 3.223{text}**{res} 0.640
(0.826) (1.650)
{text}Cut Four{res} 4.186{text}**{res} 1.651
(0.875) (1.654)
{text}
** p<0.001; * p<0.01; two tailed
{title:Notes}
This program is inspired by (and some programming lifted from) Christopher
Ferrell's {help esthold} and {help estprt} commands. Code for parsing the multiple
equation syntax was taken shamelessly from the {help reg3} command - thank you,
Vince Wiggins and Statacorp.
The tag, row number and row type columns can be used to sort the rows in
useful ways once inside the spreadsheet. For example, sorting on row type,
one would then have blocks of coefficients, standard errors, and results
grouped together, possibly across a large number of tables (different tags).
So one can easily set formatting options for each type (block) of row,
and then resort by tag and row number to return to the orginal output order.
If you set things up right, however, you should be able to take these results
directly into a word processor.
The significance calculations are based on e(df_r) containing degrees of
freedom for t ratios; for models that do not generate e(df_r) (e.g. {help probit})
the z ratio is used.
{title:Author}
Nicholas Winter
Department of Political Science
Cornell University
nw53@cornell.edu