Hi Seema,
I believe your message has been posted correctly, and I think everyone
else on the list should have received it (provided they decide to read
messages related to this particular thread).
I am pretty new to Stata myself (like you I have joined this list to
sort out some problems I've encountered myself). But it sounds like
what you really need is statistics advice (and I'm not an expert at
this either). If what you asking is what is a conventional statistic
for comparing goodness of fit across various models, other than simply
looking at percent differences in R-squared values, I'm not sure, but
no doubt others on this listserve will have some advice.
What I recommend is to re-post your question under a new thread (this
should be simply a matter of re-sending your email to the same
address, but changing the subject heading to reflect your question)
and someone should respond.
Best of luck!
Chris
On Mon, Nov 9, 2009 at 11:56 AM, Seema Pissaris <[email protected]> wrote:
> Dear Chris,
>
> I am a new user to Stata and am terribly confused.
>
> First: how do I post a message on list serve?
> Second: I need to calculate an R squared change from two panel regressions.
> What is the best way to do this? The reviewers have asked for hierarchical
> results, so I need to have incremental changes to the R square. Is this
> possible?
>
> Thanks so much for your help on this Sunday evening.
>
> Take care,
> Seema
>
>
>
>
>
>
>
> Seema Pissaris Ph.D.
> Assistant Professor
> Andreas School of Business, Office 207
> Barry University
> 11300 NE 2nd Avenue
> Miami Shores, FL 33161
> (954) 579-6707 (Cell)
> (305) 899-3530 (Office)
>
>
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Christopher
> Hajzler
> Sent: Sunday, November 08, 2009 5:47 PM
> To: [email protected]
> Subject: Re: st: Stata 10 is Malware (also version 11?)
>
> Filipa,
>
> I'm still curious: did your students discover that they had all been
> using illegal versions of the software and that this was likely
> causing the problem? And were entire variables being dropped after a
> particular operation was conducted?
>
> I'm just a little concerned that, if this type of capability is built
> into the program to prevent piracy, it could accidentally occur even
> on legally installed versions. Perhaps someone with advanced computer
> or programming knowledge would find the suggestion ridiculous, but I
> find that several day-to-day programs often don't function as they
> should (a windows user could probably guess which company's mediocre
> software I'm referring to) so it is a bit disconcerting to think that
> STATA might suddenly develop a "glitch" and drop part of the dataset
> being used.
>
> If it is merely dropping of variables, I suppose it should be obvious
> if this were to ever happen. But if it applies to observations as
> well, the potential for a glitch to occur might be cause for concern.
> It also makes me wonder if it wouldn't have been better to just design
> STATA to just not work altogether when unauthorized licenses or
> activation keys are used.
>
> Best wishes,
> Chris
>
> On Sat, Nov 7, 2009 at 2:59 AM, Filipa de Castro
> <[email protected]> wrote:
>>
>> Demo Crazy:
>>
>> I had some students having the same problem you report, some time ago,
>> which took me ages to understand what was going on and was very
>> unpleasant. But once I realized what was happening (they were 3 with
>> the same dropping variables problem) they had a mature reaction not
>> blaming STATA for losing hours of work rather they assumed they error
>> and apologized and accepted a penalty in their marking, and off
>> course, moved on to use legal versions of STATA.
>>
>> I can´t believe you really think you could sue STATA for a pirate copy
>> that does not work well.
>>
>> Filipa
>>
>>
>>
>>
>> On Fri, Nov 6, 2009 at 4:46 AM, Demo Crazy <[email protected]> wrote:
>> > After seeing this post
>> >
>> > http://www.stata.com/statalist/archive/2008-08/msg01142.html
>> >
>> > I tried to install my *legal* copy of Stata 10 with one of those serials
> available on the web. I found the same behaviour: random dropping of
> variables.
>> >
>> > Therefore the original Stata itself (not version modified by others as
> gus from Stata suggest) includes undocumented functions: this is definition
> of MALWARE. Is it including also other undocumentd parts of code that allows
> it to steal my personal data?
>> >
>> > I thinks that this behaviour from Stata can be sued. Just a notice
> telling that this dropping has beed made because the license is not valid
> would make it legal. But otherwise Stata guys are crackers, includign
> malware in their product.
>> >
>> > Thank you a lot STata: you have finaly convinced me that I won't ever
> buy a new copy of Stata and just move to other statiscal package (preferibly
> open source like GNU R, but maybe SAS or another)
>> >
>> >
>> >
>> >
>> >
>> >
>> > *
>> > * 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/
>
> *
> * 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/
>
*
* 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/
