Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Royston's mvis ado and version questions

From   "Nick Cox" <>
To   <>
Subject   RE: st: Royston's mvis ado and version questions
Date   Wed, 29 Sep 2004 10:38:55 +0100

To Richard's excellent explanation I will add a 
few comments: 

Even Stata user-programmers tend to start writing 
programs using a new version's features pretty soon as it 
is released. Sometimes that's essential for what 
they want to do; sometimes it is just more congenial 
to use improved language features. We are now some way 
into the lifetime of Stata 8 and it's a sad fact that 
those still on previous versions are increasingly 
left behind in terms of being able to use the 
full range of Stata add-ons. 

In the case of -mvis- a quick glance suggests
that you might be able to hack it to -version 7- 
and it would remain legal and correct. But that 
would gain you nothing, really. -mvis- is just part 
of a package and the whole package would need 
to be scanned in a bid to remove Stata 8 features, 
which would I guess require considerable programming 
expertise. Someone would need to able to recognise
Stata 8 features and translate them into a Stata 7 
equivalent; people with this knowledge tend to 
be using Stata 8 already. 


Richard Williams replied to Dave Purcell

> >A few questions:
> >
> >1) I just installed Patrick Royston's mvis ado (and related 
> files) and
> >tried to run it, but I'm getting a "version 8 not supported" 
> error. Is
> >it safe to assume I can't use this with the version 7 I'm 
> running (Born 08 Aug 2001)?
> >
> >3) Is there a way to tell if an ado file requires a certain version?
> >I've looked through the FAQs and haven't come across an answer.

> Stata recommends that you specify the version of Stata you 
> are using at the 
> top of programs and do-files that you write.  Hence, 
> somewhere early on in 
> an ado file you'll see a line like
> version 8.2
> Try to run that program in an earlier version of Stata and you'll get 
> errors like you saw.
> Now, it is always possible that the program would work fine 
> with an earlier 
> version of Stata.  So, you could try making a copy, changing 
> the version to 
> 7, and see what happens.  But, keep in mind that (a) it may 
> not work at 
> all, because it requires version 8 features, and (b) worse 
> still, it may 
> seem to work but give the wrong answers.  The author will 
> typically only 
> have tested it with the version of Stata they happen to have, 
> so there are 
> no guarantees if you try it with a lower level version.
> So, I suppose you could try to hack the programs; but if you did, I'd 
> suggest making sure you run them on problems with known 
> results to see if 
> it seemed to be working correctly.  I don't know anything 
> about -mvis- but 
> I see that it deals with missing data, so one immediate 
> concern I would 
> have is whether it is dependent on the changes in MD that 
> were made in version 8.

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index