|Title|| The Statalist FAQ
|Author||Nicholas J. Cox|
|Date||October 1999; updated June 2013; minor revisions June 2017|
|1.||Before you post|
|2.||How to use Statalist: mechanics|
|3.||How to use Statalist: advice|
|4.||The names Stata and Mata|
|5.||What to do if you do not get an answer|
|6.||Private emails to those active on Statalist|
|7.||FAQs on Stata|
Statalist is an email-based discussion group for users of StataCorp's Stata statistical program. It is not moderated. The list is closely followed by the list maintainer.
Statalist was started in August 1994.
The appropriate subjects for discussion on Statalist are
Postings on anything to do with Stata are welcome, such as
Once a statistical topic is mentioned, comments on the statistical or scientific merits and demerits of different approaches are legitimate. Statalisters want to use Stata to do good statistical science, and many members value Statalist for the broader material it includes.
(This is for anyone new to lists like Statalist.)
Listservers send and receive email. When anyone sends email to a listserver like Statalist, it is forwarded to all the subscribers of the list. Later, if someone feels moved to respond, that person sends email to the listserver, and that, too, is forwarded to all subscribers. Thus the process of using a list is
Some people join lists and just watch—that’s called lurking in netspeak, and, despite the negative connotation of the word, there is nothing wrong doing that. You can learn a lot by lurking.
Once posted, your contribution is archived indefinitely. There is no deleting.
Statalist is one of several lists managed by the majordomo list processor running on the Harvard School of Public Health’s hsphsun2 computer.
Majordomo handles requests for subscribing and unsubscribing from email lists. Some requests, like subscribe, unsubscribe, who, or help, are sent to majordomo not to Statalist. In reply, you will get an email message with the information and action that you want.
Questions that should be sent to all readers of Statalist are sent to statalist not majordomo.
In either case, you must communicate in plain text. See also 2.2 below.
There are two lists under the general label of Statalist: statalist and statalist-digest. The former forwards all posted messages, and the latter only a digest version at the end of the day. You may prefer the digest if you wish to receive fewer messages. It is easier to skim one long digest message than to skip through several individual messages (currently averaging about 40 each day).
The software considers statalist and statalist-digest separate lists, so when you unsubscribe, make sure it is from the correct list.
Statalist is maintained by Marcello Pagano, Harvard School of Public Health. email@example.com
The Statalist FAQ (this document) is maintained by Nicholas J. Cox, Durham University. firstname.lastname@example.org
StataCorp does not run Statalist, but StataCorp staff monitor Statalist constantly and contribute frequently.
None of these people or organizations is responsible for the content of Statalist postings, which are all informal public communications, or for the consequences of acting upon them.
Statalist is not moderated. Thus we rely upon the good sense and the good manners of Statalist members to maintain the aims of the list.
Before posting, consider other ways of finding information:
findit can tell you about not only all of the above, but also about community-contributed Stata programs available on the Internet.
From Stata 11 on, a PDF version of the manuals is included with each copy of Stata so that all users have access to the manuals.
Before you post to Statalist, please read sections 2, 3, and 4 of this FAQ. Doing so will only take a few minutes but is likely to save you embarrassment and frustration. The main aim of these sections is to provide precise advice on what does and does not maximize your chances of quickly getting answers that are clear, correct, and complete.
Only subscribers may post to the list, so you must subscribe before posting; see above.
In addition, note that your posting will succeed only if you are mailing from the same account that you used to subscribe to the list.
If you wish to subscribe with one address (to either Statalist or Statalist-digest) and to post from another, please email Marcello Pagano at email@example.com with your alternative address(es).
To send a posting to Statalist that goes out to all users,
Note: if you get the statalist-digest version of Statalist, you should still post your questions to firstname.lastname@example.org.
If you are replying to a posting, your mailer will typically produce a sensible message title automatically (e.g., by prefixing the previous title with “Re:”).
However, if you are replying to a daily digest, put your reply in a new posting, taking care to use the same title as the specific posting you are replying to. Replying to the digest itself (even if you edit out all irrelevant content and change the title) causes archiving software to receive irrelevant information about what you are replying to.
In addition, please do not start a new thread by replying to a previous posting. Again even if you delete the previous posting’s contents and change its title, such practice messes up archiving.
Because posts are not moderated, a post should normally be sent to everyone on Statalist within a few minutes of your submitting it. If you do not receive a copy of your post or—especially in the case of digest subscribers—do not see it appear in the archives, then it is likely that your post was rejected for reasons described earlier in this section.
unsubscribe statalist email@example.com
Replace firstname.lastname@example.org with your actual address.
Note: if you get the digest version of Statalist, change statalist to statalist-digest.
See also 2.7 below.
If there is a problem with your email server or your email address changes, you might generate enough error messages that you are removed from the subscriber list.
If this happens to you, resubscribe.
See also 2.7 below.
There is a digest version of Statalist. In order to receive that, you need to unsubscribe from Statalist and subscribe to Statalist-digest. Do the following:
unsubscribe statalist email@example.com
subscribe statalist-digest firstname.lastname@example.org
Replace email@example.com with your actual email address.
To receive even fewer messages, you also have the option of unsubscribing from Statalist and scanning the archives from time to time.
In order to receive the regular version of Statalist, you need to unsubscribe from Statalist-digest and subscribe to Statalist. Do the following:
unsubscribe statalist-digest firstname.lastname@example.org
subscribe statalist email@example.com
Replace firstname.lastname@example.org with your actual email address.
The archives can be found in two places:
Some people use their mailer to send out-of-office messages to Statalist when they travel. You are asked not to do that, because such messages are of no interest or use to almost all readers. If you do send messages of that kind to Statalist, you will usually be unsubscribed by the moderator. You may naturally resubscribe when you return.
Similarly, you are asked not to set up your mailer to send such messages to individual posters who send to Statalist.
One suggestion is to exclude sending out-of-office replies to any messages that contain st: in the subject line instead of filtering on the email address email@example.com. This is, however, not absolutely guaranteed to exclude only Statalist messages or even all Statalist messages.
The simplest solution of all is to unsubscribe when you go on vacation.
Address email to firstname.lastname@example.org.
Statalist is maintained by Marcello Pagano. If you have private comments or difficulty with the automatic subscription software, please contact him.
People posting on Statalist may also think about posting the same question on other listservers or in web forums. There is absolutely no rule against doing that; it is not our business to constrain what you do elsewhere.
But if you do post elsewhere, we ask that you provide cross-references in URL form to searchable archives. That way, people interested in your question can quickly check what has been said elsewhere and avoid posting similar comments. Being open about cross-posting saves everyone time.
Cross-posting does not affect the request elsewhere in this FAQ that you close threads on Statalist. If your question was answered well elsewhere, you are asked to post a cross-reference to that in a closure on Statalist.
As with all such lists, Statalist relies on the generosity and good manners of its members to function well. The following guidelines arise from reflection on what does—and what does not—ease understanding of questions and production of answers.
You might also find various websites that discuss general issues in getting help from technical lists instructive and even amusing. Mike Ash discusses “Getting answers” at http://www.mikeash.com/getting_answers.html, with key headings:
Eric Raymond and Rick Moen discuss “How to ask questions the smart way” at http://www.catb.org/~esr/faqs/smart-questions.html. We don’t endorse every detail in this statement, but we particularly underline that scorn, sarcasm, and bad language are deprecated on Statalist, although irony and more gentle wit have been observed on occasion. However, Raymond and Moen capture much of the ethos and style that characterize technically oriented lists such as ours.
As already mentioned, once you have posted something to Statalist, it appears in the archives. Any less than gracious postings will remain visible to embarrass or handicap the sender forever, so don't send them in the first place.
Stata/MP 13.0 for Windows (32-bit) Revision 07 Jun 2013
I am using -xtreg- in Stata 12.1.Example 2:
I am using -estout- from SSC in Stata 12.1. Typing -which estout- in Stata shows meExample 3:*! version 3.13 06aug2009 Ben Jann
I am using -metareg- from SJ 8-4 in Stata 12.1. Typing -which metareg- in Stata shows meFor more explanation, see 8. Ado-files FAQ below.. which metareg *! v2.6.1 Roger Harbord 4 Nov 2008 *! v1.06 copyright Stephen Sharp January 1998 STB-42 sbe23
. sysuse auto, clear . scatter mpg weightI want to attach a text label to an outlier like that at 41 mpg. How do I do that?
Stata is an invented word. Some pronounce it with a long a as in day (Stay-ta); some pronounce it with a short a as in flat (Sta-ta); and some pronounce it with a long a as in ah (Stah-ta). The correct English pronunciation must remain a mystery, except that personnel of StataCorp use the first of these. Some other languages have stricter rules on pronunciation that will determine this issue for speakers of those languages. (Mata rhymes with Stata, naturally.)
Stata is an invented word, not an acronym, and should not appear with all letters capitalized: please write “Stata”, not “STATA”. Mata is also an invented word, not an acronym.
Sometimes, a posting gets no reply. It is possible that your posting got overlooked, but this is much less likely than you might think.
If you get no answer, you might be tempted to repost the question, but please think twice before you do. The same post reappearing repeatedly strikes many Statalist members as impatient and inconsiderate. You had your chance, but, unfortunately, no one wanted to answer the question.
If your posting gets no reply, it may be the fault of the question, of the questioner, or of those who read the question. Who knows for sure? However, it is most likely to be for one or more of the following reasons:
It is possible that you may benefit from trying to make your problem much clearer or simpler. Remember that a very long posting with a mass of detailed explanation is just as offputting as a question that is cryptically brief. The best advice is to rewrite the question so that the key issue is made as clear as possible but also is stated as briefly as possible.
But in all circumstances, there is a simple rule of thumb: A rewrite or even one repost of the original is tolerable, but more than one repost is not. If after two attempts you have not received an answer, there is too slim a chance that you will get an answer on Statalist to warrant another attempt.
People very active on Statalist or in the Stata user community can get many personal emails from people they hardly know requesting Stata support. This is not very surprising: people who want help naturally tend to ask those who they think may provide it. But, for everyone’s sake, this may need a little more consideration than it is sometimes given.
In particular, some posters are tempted to reply directly to people who answer their initial questions. You are asked not to do that. Instead keep the thread public so that anyone interested can see any further discussion and so that no particular person is pressured to write further if they do not wish.
There can be many reasons to email someone directly: most obviously, if you are a friend or you have a question specifically about a program someone has written. Nothing below applies to those kinds of mail.
However, emailing someone on the grounds that they appear active and knowledgeable about Stata is a tactic with a low probability of getting precisely what you want and a high probability of wasting the time of both parties.
So, for all these reasons, it is usually better to post directly to Statalist.
The answers to the most Frequently Asked Questions (FAQs) about Stata can be found at StataCorp’s website:
These are worth reviewing periodically, even if you are not searching for a particular answer. On the FAQs page, you can click on "What's New in the FAQs" to see what has been added recently.
The FAQs are indexed in Stata’s online search but only if you install the updates. See "Updates to Stata" below.
For specific information on the Stata Technical Bulletin, point your browser to http://www.stata.com/support/faqs/resources/stata-technical-bulletin-faq/.
For specific information on the Stata Journal, point your browser to http://www.stata.com/support/faqs/resources/stata-journal-faq/.
Every few years, there is a new release of Stata from StataCorp.
A given release, however, is updated periodically. These updates are available free over the Internet.
If you have an Internet connection, the easiest way to do this is with the update command. From within Stata, type
update from http://www.stata.com
and follow the instructions. This will work with any version of Stata; Stata will not confuse updates for one version of Stata with updates for another.
Updates are also available via your browser, http://www.stata.com/support/updates.
Statalist is one way to learn about Stata. In addition, StataCorp offers courses over the Internet that range from beginner to advanced. Point your browser to http://www.stata.com/netcourse.
In addition, there are various books published that can help—visit the Stata Bookstore at stata.com/bookstore.
Programs published in the Stata Technical Bulletin (STB, the bimonthly journal, which appeared between March 1991 and May 2001) and in the Stata Journal (SJ, the quarterly journal, 4th quarter 2001–present) are available for free over the Internet.
The easiest way to do this is from within Stata. Suppose you want to download gr0001 from SJ1-1. Type
net from http://www.stata-journal.com net cd software net cd sj1-1 net install gr0001
In windowed flavors of Stata, you can do this with your mouse. First, click Help and then SJ and community-contributed programs.
Type help net for further information.
In addition to downloading programs, you will usually want to read the original articles. Electronic access is free for articles published 3 or more years ago. For information on that and on accessing newer articles, see The Stata Journal archives or Individual Stata Technical Bulletin (STB) issues.
Publications to the STB or SJ are indexed in search and findit. search will only yield up-to-date results if you install the updates periodically. See "Updates to Stata" above.
Kit Baum (email@example.com) maintains the Statistical Software Components (SSC) archive at the Boston College Department of Economics. This can be found at
http://ideas.repec.org/s/boc/bocode.html, hosted by the University of Connecticut Department of Economics
http://econpapers.repec.org/software/bocbocode/, hosted by the University of Örebro Department of Business, Economics, Statistics and Informatics
This archive covers most of what has appeared or been publicized on Statalist since September 1997. Some of it is now outdated by later additions to official Stata, and some of it is obsolete for other reasons, but a lot of it remains the best way of doing many tasks. Much of it has not appeared in the STB or SJ and may never appear there.
For official explanation and for detailed notes on how to submit to and retrieve from the archive, see http://repec.org/bocode/s/sscsubmit.html.
Most programs being added now require recent versions of Stata, but much of the older code may work under earlier versions. Recent additions, and some older ones, specify the required version.
Some authors send updates of programs directly to Kit Baum (firstname.lastname@example.org) without publicizing the updates on Statalist, especially those regarding minor bug fixes or enhancements. Hence, do not copy programs posted on Statalist from old email messages; always go straight to Kit's site.
By far the best way to interact with this site is with the official Stata command ssc. If you want information on the hotdeck module, type ssc describe hotdeck. If you want to install the hotdeck module, type ssc install hotdeck [,replace]. (ssc install, like net install, will not overwrite existing files unless replace is specified.)
Alternatively, local restrictions may oblige you to use a browser rather than ssc. Be warned, however, that your browser is not as smart as Stata in understanding Stata program files correctly, especially with regard to ends of lines and handling .hlp files. In fact, your Stata may tell you that there are syntax errors in ado-files that have been downloaded. Typically, those errors have been introduced by your browser. Similarly, Windows in particular has its own help files that postdate the introduction of Stata .hlp files.
In windowed flavors of Stata, you can do this with your mouse. Click on Help and then on SJ and community-contributed programs. See [R] net for further information.
You may also search to see what is in the SSC archive by accessing http://ideas.repec.org/s/boc/bocode.html and using the “Search this series” feature (there are several pages of listings, so a Find on this Page will not be comprehensive). Using this method, you can find tools for ‘inequality’ or those by ‘Jann’. This search covers the full text of the abstracts, as well as the titles of the items posted.
Alternatively, you may use findit to search for community-contributed material, including the SSC archive.
You may see what has been added to SSC within the last month by typing ssc new (or in Stata versions 7 to 9 ssc whatsnew). This provides clickable links for the description and installation of new and revised SSC items.
Alternatively, you may see what has been added to SSC by accessing http://repec.org/docs/ssc.php in your browser.
In addition, various users have made programs available at different sites. See stata.com/links/resources-for-adding-features for some links to other user sites.
You may use findit, which finds and lists sources of information on Stata and Stata commands that have already been installed on your computer or are available on the web. The results include (1) official help files installed on your computer, (2) FAQs available at the Stata website, (3) material published in the STB and in SJ, and (4) community-contributed programs and help files available over the web.
It is also possible to access these sources using your normal browser. A good search engine should catch material not referenced via the web pages mentioned above. However, downloading software is much easier using Stata.
Much of this material is in the form of packages that may include several files. Much of it is also available in the SSC archive.
There is no inevitable progression from the SSC Archive to the Stata Journal and certainly no inevitable progression from either to official Stata. These are operated as separate, but not isolated, entities. Many users post programs first in the SSC Archive; they may or may not then publish in the SJ. That depends on whether they submit to the journal and on the editorial process of the journal. Some users who publish in the SJ make little or no use of Statalist. Those are matters largely of personal habit.
Community-contributed programs accessible from the StataCorp website have essentially the same status as those available from the SSC Archive. Their posting there is basically a matter of convenience to users and is not an official endorsement by StataCorp.
However, StataCorp has nonexclusive rights to any program published in the STB or SJ, while anything placed in the SSC Archive is tacitly put in the public domain. In practice, you can probably take anything published in either medium and modify it as you will—especially if you do that privately—but publicly we recommend that, unless you are the original author, you change the name of the program, take all blame for any limitations your changes produce, and imply that a suitably large portion of the credit for the program belongs to the original authors.
StataCorp is entirely responsible for decisions about what is incorporated in official Stata.
Generally, the answer is yes. An ado-file with the same name as a previous version will normally correct bugs or augment the functions of the superseded version.
The tag (in the STB the insert number) indicates the version of the file. Suppose st0067 provides some feature. st0067_1 should provide all the same features and be better than st0067. st0067_2 should be the same thing but better than the two that preceded it, and so on.
We say “generally” only because authors and editors sometimes make mistakes. In terms of installation, if the name of the command is the same, then the new automatically replaces the old.
Incidentally, make sure you are aware of the scope for using the adoupdate command to keep your installed packages up-to-date. See [R] adoupdate.
It cannot, in the sense that no community-contributed material could change the behavior of an official command of Stata such as regress or logit. An STB/SJ contribution, however, could offer an alternative to an official command, such as offering a command named regress2 or logit2. In that case, the command should do whatever the author says it does. In most cases, a numeric suffix would mean that the author is proffering this as a replacement for the original, but there are no rules.
Stata’s search database tries to keep track of that information. When StataCorp incorporates a feature from a previous STB/SJ submission in its entirety, it marks the original insert as historical.
For instance, type search glm, and you will get a list of suggestions that includes previous STB/SJ submissions. All those suggestions should be relevant to present users. Type search glm, historical, and the list will include more suggestions–all the above plus more STB/SJ submissions. The extras are not really relevant now because those features have been incorporated into Stata.
We can turn this around. Say that you stumble across sg16.1 somehow. You can look up items in the search database by their reference number by specifying the entry option. If you type search sg16.1, entry, Stata should tell you about the sg16.1 insert. Stata, however, will say nothing. It is as if sg16.1 does not exist.
If, on the other hand, you type search sg16.1, entry historical you will learn that sg16.1 was published in STB-12, that its title is "Generalized linear models using power links", and that it was written by J. Hilbe. search with the historical option looks in entries of historical interest, too. sg16.1 is interesting, historically speaking, but not of practical use today because Stata now has a glm command.
But note there is absolutely no guarantee that StataCorp will incorporate any contribution from the STB/SJ into official Stata. In the long run, all users benefit because Stata is centered on a stringently selected core of code that is both of high quality and of use to many, if not all, users.