apparently i only thought i’d estimated a model

Major thing off my list for this sabbatical is revising the book on categorical dependent variables in Stata that I have with Scott Long. (No word yet on when it will be out.)

I might say more about that project later, but just wanted to mention a style tidbit from working with our editor at Stata Press, who’s both a great econometrician and great guy.* He took us to task for all the times in our draft we made reference to “estimating a model” (or, e.g., “we estimated a logit model in which…”). We were offered two alternatives: saying that we “fit a model” or that we “estimated the parameters of a model.”

Given that I see “estimate a model” phrasing all the time, I thought I’d pass it along to anyone else who may be as eager as me to maximize the extent to which we at least look like we know what we’re talking about.

* Bonus fun fact: even though this person has been at Stata since before the first edition of our book, and so is by no means young, his dissertation adviser was my colleague Art Stinchcombe’s son. Small world, academia.

Author: jeremy

I am the Ethel and John Lindgren Professor of Sociology and a Faculty Fellow in the Institute for Policy Research at Northwestern University.

3 thoughts on “apparently i only thought i’d estimated a model”

  1. This is a timely warning: I am in the midst of the copy editing for the (much larger) second edition of Counterfactuals and Causal Inference.

    Although I understand the point of your editor, I don’t think ‘estimated a model’ and ‘fit a model’ are all that different, at least for 99.9% of readers we are all trying to reach. I have used both and will continue to do so.

    Of course, “estimated the parameters of a model” is correct, but it also leaves the impression that someone else chose the specification, that you just made up the model chi-squared, and so on. Certainly, you wouldn’t want to have to clarify that, yes, you too did all of that. As a result, I think “We estimated five logit models, and Table 5 presents …” seems a good way to throw the reader into our customary tour through the features of a bunch of things we have produced by issuing commands to a computer.

    Rather than worry about this, I’d rather we spend more time banishing *s from our tables, which is our discipline’s chosen “chartjunk.”

    In any case, I very much look forward to your revised book, so that it will save me from having to relearn the syntax of “margins” in Stata every six months. Your Stata editor should perhaps spend some time on that, asking his colleagues not to use things like “over,” “at,” and “if” in ways that require one to consult the Stata manuals over an over again to remember which one does what. That would be a great service, while we await your book and its customized commands.

    Like

  2. Yeah, various side points in your comment could inspire three posts. As to your main point, to be sure, the difference between “estimate a model” and “fit a model” does not seem large to our ear, either, as we had used the former many, many times, but it was clear from discussion that it was nails-on-a-chalkboard for our editor. I was reminded of a student who learns for the first time that there are people out there who really dislike “impact” as a transitive verb.

    Like

  3. Ha. I get a mild nails-on-chalkboard feeling when I read “data is” in the Stata manual.

    “Estimate a model when your data is…” seems only fair.

    Like

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s