[JAE Logo]

Journal of Applied Econometrics Data Archive

Instructions for Authors

Authors of all papers published in the JAE are required to submit the data they used, unless they are confidential (see below). Authors are also encouraged to provide whatever other material is needed to ensure that their results can be replicated without excessive difficulty. This might include computer programs or technical appendices that are not part of the paper itself.

Before attempting to send me any datasets, please send me an e-mail message explaining how the data are organized and roughly how much space they take up. This step can be omitted if you are confident that the both the readme file and the data file(s) meet the standards of the Data Archive. My e-mail address is jgm@econ.queensu.ca. The body of the message could be essentially a draft of the "readme" file that must accompany your data. Please start with bibliographical information in the following format:

   John Smith, "An Econometric Analysis of the Price of Plutonium", Journal of Applied Econometrics, forthcoming.

All datasets must be accompanied by a readme file, a plain text file in ASCII format. This file should provide all the information needed to make use of the data. In particular, it should explain how many observations there are, what each variable is, how the data are organized in the data files, and the sources from which the data were taken. See the Data Archive for examples.

All data should be stored in plain ASCII (text) files that can be read on any computer system. The use of CSV files is encouraged. The data may be stored in one or several files. Proprietary binary formats, especially ones that are not portable across machines (such as Stata .dta files or SAS datasets) are not acceptable by themselves. However, they may be provided in addition to the ASCII files if this would make it easier for many users to access the data. To avoid problems when reading the data, non-numeric characters, if any, should occur only at the end of the file or at the end of each line; this does not apply to CSV files.

For most data sets, the simplest approach is probably to send them to me as MIME attachments to e-mail. If the files are no more than 80 columns wide, it should be possible to send them without MIME encoding. In this case, make sure that the beginnings and ends of data files are well marked. If there are several files, or any of the files is large, all of the files should be combined into a single zip or rar file.

For extraordinarily large datasets, it may be necessary to use ftp, scp, or an intermediate host such as Dropbox. It will be necessary to make special arrangements to handle such cases.

Supplementary material, such as computer programs and data appendices or technical appendices that are not published in the journal itself, should also be sent to me. Appendices should be in the form of PDF files.

Confidential data:

Some data sets are confidential. Please check with the source of the data if you are not sure whether they may legally be made available on the Web.

When the data are confidential, we ask authors to provide a readme file that provides a reasonable amount of information about the data. In particular, the source of the data must be described in enough detail so that other researchers can apply to obtain access to them. As examples of what is appropriate, please see the following:

It is highly desirable, as in the case of the second example, to provide the programs that were used to extract the data from the original source files.