Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down at the end of May, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Nick Cox <n.j.cox@durham.ac.uk> |

To |
"'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu> |

Subject |
RE: st: New package -use10save9- available on SSC |

Date |
Wed, 2 Nov 2011 19:15:55 +0000 |

Thanks for the reply. I wonder whether you are confusing quite distinct issues. When you get access to Stata 12, you will clearly be able to read Stata 12 files and use -save, old- to make them available to Stata 9 (among other versions). Also, whether your program requires Stata 12 is not the question. It does not: it requires version 9 and should therefore run in Stata 12. What is now my main question is completely different, namely whether your program can allow Stata 9 users to read Stata 12 files, which I take to be implied by the claim that "use10save9 is a program that helps out with opening Stata 10+ files from within Stata 9 and then saving them in Stata 9 format (using use10 and save9)" if and only if "10+" is interpreted to include 12. If you don't intend to imply 12, you just need to fix your help file. Otherwise the way to answer this is to test it, and I have to say (again) that I can't see how you can do it, but a simple refutation by showing me that you can do it will oblige to say very publicly that I am mistaken. A crucial detail is that -use10- (SSC), which you use, was written in 2008 and so necessarily quite unaware of later changes to Stata's dataset format. Nick n.j.cox@durham.ac.uk Lars Ängquist Dear Nick. Many thanks for your comments. As you indicate the primary use of -use10save9- should be, and was planned to be, to facilitate for the Stata 9 user to in a one-command way being able to transfer multiple files from e.g. Stata 10 format to a format that would be readable for such a user. (An aside is that this was initialized by an actual need, from one time to another, for some Stata users at the Institute where I work, where in fact still quite a few users are using Stata 9.) Initially I designed the program to simply use Stata's built-in -save- function when, after each loading of a matched file, re-saving the file and hence to make it useful strictly from Stata 9-only, but then I noticed on Statalist that (an update to) -save9- was made available and hence thought that I could perhaps add a secondary application addressing this task from, so to say, the other way around in the sense of facilitating a one-command option to, from e.g. Stata 10, calling the program in order to save a matched set of files in an older-than-used format (instead of loading, and re-saving, files having a more recent-than-used format). Admittedly, this task should also be directly solvable by usage of the in-built duo -save- and -saveold- but I here considered this approach to be quite convenient in the sense of not having to check the user-version; since e.g. -saveold- and -save- then appeared to be appropriate for the task from Stata 10 and Stata 9 respectively. (Another aside, originally I thought that -saveold- used from within Stata 12 would not save data in a format readable from Stata 9 since the Stata 12 format was described as a new format, and hence here introduce further complications in this sense, but I later found out that I was probably mistaken in that case, see e.g. http://www.stata.com/support/faqs/data/commaold.html.) Regarding the usage of -use10- and Stata versions, the command at least seem to be working for Stata 11.2 (which I currently run) and my only defense in this respect is that I, admittedly somewhat naively, assumed that it would perhaps also work in Stata 12 based on the information stated in the help-file that the program will in fact use the standard -use- in cases of not recognizing the Stata 10 format and hence then assumingly work also for these cases. I have unfortunately not had the possibility to test this myself in Stata 12, since it is currently not available to me, but I would be very grateful if someone did that and - if not working properly - I would happily update the -use10save9- help-file in order to state that it then requires Stata 9/10/11. I do not claim any other functionality of -use10save9- than the primary and secondary ones outlined above (except for some limited side-effect possibilities of renaming multiple files, but for such purposes I would perhaps recommend another of my functions -renfiles- available on SSC). With best wishes & regards / Lars Ängquist --- --- --- Date: Mon, 31 Oct 2011 09:15:43 +0000 From: Nick Cox <njcoxstata@gmail.com> Subject: Re: st: New package -use10save9- available on SSC This program raises questions similar to, but not identical to, those raised by -save9- (SSC). See earlier postings by Joerg Luedicke http://www.stata.com/statalist/archive/2011-04/msg00654.html http://www.stata.com/statalist/archive/2011-10/msg00928.html and myself http://www.stata.com/statalist/archive/2011-10/msg01158.html http://www.stata.com/statalist/archive/2011-10/msg01253.html Marco Ercolani's replies are included in these postings. (He is the author of -save9-.) First, -use10save9- is I gather essentially for users of version 9, as if you have Stata 10 up you don't seem to need it (except for its support for multiple files). That is clearly a class of users who might seek some help in reading .dta files that otherwise are unreadable. However, if you have version 9, then inbuilt -save- will automatically save to version 9 format. So, the use of -save9- within -use10save9- appears superfluous. Second, -use10- (SSC) itself claims only that it can read Stata 10 .dta files in Stata 9, whereas the claim made on its behalf here is that you can read Stata 10+ files. This seems all too likely to be misunderstood by some users, as Stata 10+ surely includes Stata 12. Third, you can go beyond what official Stata provides if you write extra code that makes use of StataCorp's published specification of .dta format. -use10- does do this among the three programs mentioned but otherwise I don't see that the programs can do anything radically different from what -save- and -saveold- do in any particular version of Stata. The underlying principles are very simple. Everyone benefits from being totally precise about what programs can and, by extension, cannot do. Similarly, superfluous programs do no harm, except that they can confuse. Those are the principles. In practice -use10save9- does offer functionality for working with several .dta files at once, but otherwise I am unclear about its precise rationale. Perhaps others too would appreciate some clarification here, or correction of what I am not understanding. Nick Nick On Sun, Oct 30, 2011 at 10:29 PM, Lars Ängquist <lars.angquist@telia.com> wrote: > Dear Statalist. > > Many thanks to Kit Baum who recently (2011-10-29) made the -use10save9- > package available for download from SSC. In Stata, use the -ssc- command > to > do this directly, or go somewhat more indirectly through usage of > the -findit- command. > > --- --- --- > > Brief abstract: > > use10save9 - is a program that helps out with opening Stata 10+ files from > within Stata 9 and then saving them in Stata 9 format using the > user-written > -use10- (Radyakin, 2008) and -save9- (Ercolani, 2011) functions. May be > applied to a quite generally defined/matched set of Stata datasets within > a > defined folder and, if selected, also with respect to all corresponding > subfolders. The simplest syntax is purely based on defaults. All options > are > optional. > An alternative, secondary, usage might be to simply use the function as > an > extension to -save9- in order to save matched sets of datasets in Stata 9 > format while being in Stata 10+. * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**References**:**Re: st: New package -use10save9- available on SSC***From:*Lars Ängquist <lars.angquist@telia.com>

- Prev by Date:
**RE: st: RE: Dropping Duplicates that Aren't Exactly Duplicates** - Next by Date:
**Re: st: negative binomial model in stata** - Previous by thread:
**Re: st: New package -use10save9- available on SSC** - Index(es):