Thought Leadership
Historical Financial Statistics  
HFS Home |  Data |  Documentation |  Submitting Data |  Links |  About HFS

Submitting data

Send all data or questions to the contact on the “About HFS” page. We welcome submissions of data. They should be in Microsoft Excel. Accompanying remarks should be in Microsoft Word. If you have data in a format you cannot easily convert to Excel, ask us about conversion.

Data we seek:
Our target range of coverage is 1492 to the present. We welcome inquiries about including data before 1492 where appropriate. Data can range from intraday to annual, depending on the topic, and can include any existing or defunct country.

We seek the highest quality, most frequent data available. Data may come from any appropriate source: government archives, business records, newspapers, official documents, statistical yearbooks, scholarly publications, etc. If a data series already exists in Historical Financial Statistics but you think you have better data, submit them. We will carry multiple versions of the same series where they add significant information. For example, sometimes it is useful to know an originally reported figure, even if it was wrong, because it influenced market perceptions in ways that later corrected figures did not. Or, as in the case of GDP, there may be more than one method for estimating a figure.

We prefer submissions to be in English, but will accept them in Dutch, French, German, Italian, Portuguese, Romanian, or Spanish, of which we have some reading knowledge. Notes that are not in English will be translated into English by machine and may suffer a loss of accuracy.

Data we currently exclude:
As the “Data” page says, we currently exclude coverage of certain kinds of data because of our limited time and resources. If you wish to volunteer to supervise additions of data in those areas, contact us.

A data set, not a data archive:
Historical Financial Statistics is a data set, not a data archive. If we obtain new data that we think are better than the old data, we typically delete the old data. Exceptions include cases mentioned a few paragraphs above. To archive your data, there are many sites such as EH.net or the Inter-University Consortium for Political and Social Research. It would be wonderful if a central archive for historical financial data existed, but that is a task for somebody with more spare time than we currently have.

Copyright:
By submitting data, you affirm that, to the best of your knowledge, their inclusion in Historical Financial Statistics would not violate copyright restrictions. If you submit copyrighted data, explain in the accompanying e-mail why you think reproducing the data would be permissible. Some copyright holders allow data to be used freely provided proper credit is given. Limited use of copyrighted data is also permissible under “fair use” provisions.

By submitting data, to the extent that you may have any copyright claim, you give Historical Financial Statistics nonexclusive permission to make the data available freely for noncommercial use. Unless you specify otherwise when you submit data, you also give Historical Financial Statistics permission to make your data available to commercial users. We may, for instance, allow a commercial service to copy some of our data in return for being allowed to copy some of theirs. Historical Financial Statistics is noncommercial, so the purpose of such a transaction, if undertaken, would be to increase the data available to users, and would involve no monetary gain for the organizers of Historical Financial Statistics.

Submitters may withdraw their data by notifying us.

Holders of copyrighted data who think their copyright has been violated should contact us (see the “About HFS” page).

Information to accompany data:
To the extent possible, persons submitting data should provide the following information.

  1. Dates of the data:
    The spreadsheet you submit should make it obvious which data points accompany which dates. Also, state whether data are for the end of a period, another point during the period, period averages, or something else.
  2. Source of the data:
    If the source is an archive or business records, give the file number or other information sufficiently detailed that another researcher could readily find it. If the source is a periodical, give the author’s name (if any), title of the article, title of the periodical, volume, issue number, date of publication, page numbers, and table numbers from which you took the data. If the publication is a book, give the author's name, title, place of publication, publisher's name, year of publication, and page numbers from which you took the data. You can look in the online catalog of the U.S. Library of Congress or in WorldCat for full citations for books. If the source is a Web page, give the title at the top of the page, the Web address of the page, and the date you downloaded the data.
  3. Original source, if different from (2):
    For example, you may have obtained data on bank deposits from a government statistical yearbook, but the yearbook may indicate that the data are from monthly reports submitted to the ministry of finance.
  4. Indication about your reproduction of the data:
    It is most desirable to reproduce the data as the source has it. If you have not done so, though, how have you changed the data? Have you rounded numbers? Have you converted fractions or nondecimal currency amounts into decimals? Have you omitted data, such as listing only the daily offer price rather than the bid price also?
  5. Notes:
    If you wish to offer remarks about the sources, characteristics, or methods of gathering or calculating the data, your opinion of its quality, etc., do so in an accompanying file in Microsoft Word. The Word file can also give references to any working papers or published articles about the data that you wish to cite.
  6. Your identifying information:
    Name, e-mail address, and institutional affiliation if any.

Submitted data are subject to review and approval by an editor before being posted. Review may take up to two months. The editor may reject data if they simply duplicate data we already have; if the quality seems low; if there are copyright concerns; or for other appropriate reasons. The editor may also standardize data according to the conventions of Historical Financial Statistics, but if there are important differences between the raw and standardized data, we will also post the raw data.

If you want to help collect data:
If you want to help collect data not yet in the data set, contact us. We have many ideas about where to obtain missing data. We can tailor them to fit your interests, time available, and location. Especially if you live near a capital city, where official documents are stored, there are likely to be many sources of data that have not yet been digitized.