Re: st: How to re-program -save9- for Stata 13?

[Stas Kolenikov <skolenik@gmail.com>]

Right, SPSS does have some sort of machine learning/data mining
toolkit. Their business model though is that of base SPSS for a base
price, plus anything extra at an extra price, and I am pretty sure
that the CART module would have to be purchased separately.

-- Stas Kolenikov, PhD, PStat (ASA, SSC)
-- Senior Survey Statistician, Abt SRBI
-- Opinions stated in this email are mine only, and do not reflect the
position of my employer
-- http://stas.kolenikov.name



On Tue, Oct 15, 2013 at 11:44 AM, Nick Cox <njcoxstata@gmail.com> wrote:
> Thanks, Stas. I imagine your answer summarises well the social and
> personal reasons behind many choices.
>
> There are always plenty such. For example, I don't use R over Stata
> because I invested far too much time, energy and affection in learning
> Stata to want to throw that away, long before R was evident.
>
> On the more technical aside, one user privately pointed to
> classification and regression trees as a strength of SPSS.
> Nick
> njcoxstata@gmail.com
>
>
> On 15 October 2013 17:26, Stas Kolenikov <skolenik@gmail.com> wrote:
>> Nick, it's a matter of habit, for the most part, I believe. It has
>> traditionally been used by social scientists who collect their data,
>> as it seems to have some sort of a data collection toolkit (which I
>> have never used). People like the GUI... and you have even save the
>> syntax from your point and click operations for later use! So that's
>> as far as the strengths go. Some of the tabulations can be adjusted
>> for a prettier output more conveniently than in Stata, though.
>>
>> Your question sounds a lot like "what drives people to use Stata if
>> they also have access to R?" that we both have heard a lot on
>> StackOverflow ;). So that's a disciplinary divide. A machine learning
>> blogger would insert a snippet of R code into their post to clarify
>> what they want to say without much thinking whether a different
>> software may exist. An economist would insert a snippet of Stata code,
>> as that is the language economists think in. Along these lines, public
>> opinion researchers think in terms of SPSS syntax.
>>
>> -- Stas Kolenikov, PhD, PStat (ASA, SSC)
>> -- Senior Survey Statistician, Abt SRBI
>> -- Opinions stated in this email are mine only, and do not reflect the
>> position of my employer
>> -- http://stas.kolenikov.name
>>
>>
>>
>> On Tue, Oct 15, 2013 at 10:19 AM, Nick Cox <njcoxstata@gmail.com> wrote:
>>> Not to knock SPSS, although for many that qualifies as an acceptable
>>> sport, but what drives people to use SPSS if they also have access to
>>> Stata?
>>>
>>> I am not trying to be ironic here. It is (literally) decades, i.e.
>>> sometime last century I think, since I last used SPSS, so I am quite
>>> out of touch on its relative strengths.
>>>
>>> Please take as understood that some users prefer its user interface.
>>>
>>>
>>>
>>> Nick
>>> njcoxstata@gmail.com
>>>
>>>
>>> On 15 October 2013 16:05, Sergiy Radyakin <serjradyakin@gmail.com> wrote:
>>>> Stas,
>>>>
>>>> Stata 8 and 9 share the same data format (specification 113), so
>>>> although SPSS says it writes in 5-8, it should equally read and write
>>>> 9 (perhaps the menu text was designed at the time Stata 9 did not
>>>> exist yet). Here is another reference it supports 9:
>>>> http://www.ats.ucla.edu/stat/mult_pkg/faq/fromStata_toSPSS.htm
>>>>
>>>> But introduction of reading Stata dataset capability was apparently
>>>> added in version 14 (contrary to the above reference):
>>>> http://pic.dhe.ibm.com/infocenter/spssstat/v20r0m0/index.jsp?topic=%2Fcom.ibm.spss.statistics.help%2Fwhatsnew_14.0.htm
>>>>
>>>> Furthermore, it seems the request for support of newer Stata formats
>>>> was requested still in the 17th generation of SPSS:
>>>> http://www-01.ibm.com/support/docview.wss?uid=swg21477575
>>>>
>>>> Best, Sergiy Radyakin
>>>>
>>>>
>>>>
>>>> On Tue, Oct 15, 2013 at 9:41 AM, Stas Kolenikov <skolenik@gmail.com> wrote:
>>>>> Sergiy,
>>>>>
>>>>> regarding 2), I have access at work to SPSS 21 that seems to work with
>>>>> Stata 8 files (at least their internal "Save As" gives options from
>>>>> Stata 5 through Stata 8). From Stata 12 that I have at work, the
>>>>> -save-d files (under version 12) cannot be read by SPSS, but -saveold-
>>>>> produces the version-8-compatible data set that SPSS can work with.
>>>>>
>>>>> -- Stas Kolenikov, PhD, PStat (ASA, SSC)
>>>>> -- Senior Survey Statistician, Abt SRBI
>>>>> -- Opinions stated in this email are mine only, and do not reflect the
>>>>> position of my employer
>>>>> -- http://stas.kolenikov.name
>>>>>
>>>>>
>>>>>
>>>>> On Tue, Oct 15, 2013 at 7:46 AM, Sergiy Radyakin <serjradyakin@gmail.com> wrote:
>>>>>> Hello Marco,
>>>>>>
>>>>>> could you please explain a bit, why you see a need for such a program?
>>>>>> If you have Stata 13, this means you can use -saveold- to save in the
>>>>>> previous version, then following this FAQ all the way until version 5
>>>>>> if necessary:
>>>>>> http://www.stata.com/support/faqs/data-management/save-for-previous-version/
>>>>>>
>>>>>> If you don't have Stata 13, then you use -use13- to load the data into
>>>>>> Stata 10 or later.
>>>>>> http://radyakin.org/transfer/use13/use13.htm
>>>>>>
>>>>>>
>>>>>> So I see potential clients as:
>>>>>> 1) people having Stata 9 and no access to newer versions;
>>>>>> 2) people having non-Stata software that can read dta format of 9, but
>>>>>> not newer and the software is not going to be updated, and there is no
>>>>>> access to Stata 10+
>>>>>>
>>>>>> Regarding 1) I was told the # is pretty small, but I don't know what
>>>>>> 'small' really is. Regarding 2) I am interested in what is out there
>>>>>> (regardless whether maintained or not). Can we have a headcount of the
>>>>>> programs that directly read Stata datasets? Let me start:
>>>>>> ADePT v.5.50.5001.23617 (supports data from Stata 8 to 13)
>>>>>> ADePT Map v2 (supports data from Stata 8-11)
>>>>>> R (with package foreign) (supports data from Stata 5 to 12)
>>>>>> SPSS v.15+ (supports data from Stata X?X to 9?)
>>>>>> Stat/Transfer (supports data from any version of Stata)
>>>>>> etc.
>>>>>>
>>>>>> Regarding the implementation it is not at all difficult: the current
>>>>>> code of use13 and use10 contains all the necessary procedures to
>>>>>> implement save9 which I can do if there is any significant demand.
>>>>>>
>>>>>> Best, Sergiy Radyakin
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Mon, Oct 14, 2013 at 8:54 AM, Marco Ercolani <m.g.ercolani@bham.ac.uk> wrote:
>>>>>>> Dear Statalisters,
>>>>>>>
>>>>>>> I am asking for suggestions on how I can re-program -save9- so that it will work in Stata 13.
>>>>>>> In other words, so it will save data in a Stata 9 format from within Stata 13.
>>>>>>> I can think of two possible strategies to achieve this within Stata 13:
>>>>>>>
>>>>>>> Strategy 1: Edit the contents of the dataset so they conform to Stata 9 protocols, save the file,
>>>>>>> then edit the the file header so that it appears to have been saved by Stata 9.
>>>>>>>
>>>>>>> Strategy 2: Use an existing save facility in Stata 13 that would allow me to impose limits on the
>>>>>>> data so it respects Stata 9 protocols. Does such a save facility exist?
>>>>>>>
>>>>>>> Any help on the above two approaches or helpful suggestions for a "third way" would be welcome.
>>>>>>>
>>>>>>>  Marco Ercolani
>>>>>>>  Department of Economics
>>>>>>>  University of Birmingham
>>>>>>>
>>>>>>> *
>>>>>>> *   For searches and help try:
>>>>>>> *   http://www.stata.com/help.cgi?search
>>>>>>> *   http://www.stata.com/support/faqs/resources/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/faqs/resources/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/faqs/resources/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/faqs/resources/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/faqs/resources/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/faqs/resources/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/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/