When I saw the new Stata 10 software for Stata Corp. I couldn't believe how much faster it was. Moreover, it seems that the minimum requirements are given just for the sake of it since it works fine on my old T22 machine. Moreover, the slick look, expanded options make it even more appealing. But...hold your horses. Quite often I find out strange error messages and variables suddenly dissapearing from the dataset. So far I wasn't able to figure out what is going on but I will look into it soon. It is weird also because I never had this problem with Stata8 or 9 SE.
21 comments:
Hi, I´m having the same problems about missing variables and strange error codes in Stata 10SE, did you find any cure?
The same thing has happened to me. I am working with four databases, each one having more than 70 thousand rows and 10 or more variables, with lots of daily data, and when I need to collapse by day to obtain daily time series it turns out that some variables disappear and I receive strange error messages. The same thing occurs when I try to do graphs.
If you find out what is going on, please let us know.
Unfortunately, NO CURE so far.
On one hand I am "happy" I am not the only one with such unfortunate events, on the other what the heck is the community and all the Stata people doing? I haven't seen any complaints anywhere so far. and YES, everything is up-to-date.
The only thing I was able to do about it was just to use 9.2 or the new EVIEWS which has some nice new features.
Glad you noticed this. I'm having the same exact problems with Stata 10. It jumps out at me every once in awhile - "variable (insert var name here) not found". It's strange, because it's an inconsistent behavior...
Maybe the other versions (MP, Intercooled etc) do not have this issue. There is something to look into...
I have the same problem with my Stata 10 SE. I installed the plug provided by Microsoft: 230699_ENU_i386_zip, which is supposed to solve the problem of memory conflict between XP and Satat 10. But it does not work at all!!! I still get the message: variable xxx not found.
Any new suggestions?
I am having the same problem in Stata 10/SE, especially in graphics. I have all the variable labeled, encoded, and still it says "Variable __00006 [or other temporary variable name] not found"
There is no sign of the temporary variable and I've converted/imported all as I have done with many earlier versions of stata and stat transfer.
I have the same problem and again no cure. Did somebody try to contact the company producer? If somebody finds a solution, please share it. Thanks
I am new to stata 9.2 and I am working on progression of a disease over time I have created the data in editor but I am wanting to use it with another folder that I have set up. I have patient ID on all sets of data and I want to make sure they match up properly. from one data set to another. I know that you can copy and paste the data but I am not sure the if this will match up the ID numbers.Can you help.
Thanks
To stressed:
The easiest way would be by using StatTransfer, which does a nice job in converting .dat files to many other formats. You can get a trial here:http://www.stattransfer.com/. Hope this helps.
Hello Soren, glad i found this tiny post on Google. I am experiencing the exact same issue: when i try to reshape wide a long database, it randomly drops a variable telling me that "variable (varname) not found".
I am sure that there are no database or command errors that i am making as i am directly referring to a do file prepared by Baum (i am not running the do file, i am just copying the command from there) for "an introduction to modern econometrics using stata"
Did you manage to find any cure for it? Is even Stata-Corp avare of the existence of this problem?
Has anyone found a solution to this problem? I'm also having the same problem and it's very frustrating!!!
I'm also having the same problem with two installs (Vista and XP) of Stata 10 SE. I thought 10.1 might fix it, but it doesn't. It's been almost a year since the OP here, and still I can't Google up a solution. Is there something I'm missing? This problem is really frustrating...
Are you guys using official Stata 10 copies?
check this out:
http://www.stata.com/statalist/archive/2008-08/msg01142.html
I also had the same problem of randomly missing my variable. Later, I found that it only happens to those unofficial intalled Stata.
Ok, so I admit I wasn't using an 'official' Stata copy, but half of my data is GONE! I know I didn't accidentally type the command: drop if year==2007 | year ==2008, so whatever patch Stata has written to help it crack down on unofficial copies (which is fair), I believe they have crossed the line by taking my personal property from me (yes, that data cost money!). I guess there's no way to prove it, but I am infuriated. Can't they think of something less intrusive than deleting personal property to crack down on these unofficial copies??
Of course you are not using Stata 10 legal edition... The problem that is happening to everyone is because it is a pirated edition. There is no way to fix it, unless you buy it.
I have a legal copy of Stata, but just tried one of those "serials" available on the web. I found the same problem.
It is true that those using ilegal copies are not doing the right thing, but Stata is acting against the regulations.
It is clear that STATA 10 is MALWARE, since it includes undocumented behaviour. 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 be legal. But otherwise they are crackers, includign malware in their product.
It is time to think about moving to other products like GNU R!
Seems that Stata10 still draws a lot of attention. Indeed, dropping observations and loosing control of what is in and what not is EXTREMELY frustrating. A couple of things are clear: 1. if you are experiencing this, you have a pirate copy; 2. it might be difficult (if not impossible) to fix it. 3.buy it or get it legally (IC is decently cheap) if your dataset doesn't go beyond several thousands obs.
Another reason Stata might say 'variable (varname) not found' when the variable exists is because the variable that it says is not found is listed twice. It could also be spelled incorrectly or improperly capitalized. This is even in the official version.
I installed from an official stata disk and product key and i'm still having this issue but only when i execute from a do file. the same exact command executed in the command bar works fine. also experiencing some odd errors like "variable save not found" when i try to save...
Post a Comment