Statalist


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

Re: st: RE: "no; data in memory would be lost" warning after preserve?


From   Gabi Huiber <[email protected]>
To   [email protected]
Subject   Re: st: RE: "no; data in memory would be lost" warning after preserve?
Date   Thu, 3 Dec 2009 14:26:47 -0500

Of course. Thank you. I should have read the friendly help viewer.

xtdata, fe

attempts a fe transformation of the entire data set. It works when the
said data set consists of nothing but the numeric variables that go
into the model. Mine has other gunk in it, some of it non-numeric.

xtdata <varlist>, fe

with <varlist> restricted as appropriate, worked fine.

Thanks again,

Gabi

On Thu, Dec 3, 2009 at 1:59 PM, Nick Cox <[email protected]> wrote:
> In addition, r(2000) can mean that a string variable inhibits something intrinsically numerical. Or that there are no non-missing values.
>
> Nick
> [email protected]
>
>
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf Of Gabi Huiber
> Sent: 03 December 2009 18:56
> To: [email protected]
> Subject: Re: st: RE: "no; data in memory would be lost" warning after preserve?
>
> Thank you, Nick, for the quick help. I'm still missing something.
>
> I decided that I didn't care, for this exercise, whether data in
> memory would be lost. So I threw out the "preserve" part and I
> attempted two things, with data properly xtset ahead of time:
>
> 1) xtdata, fe
> 2) xtdata, fe clear
>
> Their outcomes were:
>
> 1) no; data in memory would be lost r(4); -- fair enough
> 2) no observations r(2000);
>
> This is strange, because the observations are all there, I can see
> them: 3660, for a strongly balanced panel with t[1:30]. Usually I get
> this r(2000) error when the data set in memory is empty.
>
> Gabi
>
> On Thu, Dec 3, 2009 at 1:30 PM, Nick Cox <[email protected]> wrote:
>> My guess is that -xtdata- has no idea that you have just typed
>> -preserve-.
>> I don't think -preserve- has any side-effect in terms of setting a flag
>> that Stata can now be destructive willy-nilly.
>>
>> Why should it?
>>
>> -xtdata- is just being conservative in your best interests. You just
>> need to spell out that you know that you are being sensible. Specify the
>> -clear- option.
>>
>> (I answered this by looking at the help for Stata 11, but it's likely
>> that the answer is the same for Gabi.)
>>
>> Nick
>> [email protected]
>>
>> Gabi Huiber
>>
>> I have a balanced panel of weekly lottery ticket sales.
>>
>> I am trying to do a bit of exploratory analysis, and I attempted a
>> within scatterplot as done in Micreconometrics using Stata, by Cameron
>> and Trivedi, Stata Press 2009, at p. 243. But I get a strange error
>> message:
>>
>> . preserve
>> . xtdata, fe
>> no; data in memory would be lost
>> r(4);
>>
>> Why would I be warned about data loss after preserve? This is Stata
>> 9.2 MP. I would appreciate any ideas.
>>
>>
>> *
>> *   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index