Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: tempvar versus tempname


From   Nick Cox <[email protected]>
To   "[email protected]" <[email protected]>
Subject   Re: st: tempvar versus tempname
Date   Wed, 30 Oct 2013 00:20:22 +0000

... which is guaranteed ...
Nick
[email protected]


On 30 October 2013 00:15, Nick Cox <[email protected]> wrote:
> -tempvar- is, as I understand it, part psychology, part syntax.
>
> You are just asking for a temporary name and indicating to whoever
> reads the program, not least yourself, that you _intend_ to create a
> variable and use that name. If you lie, and never create that variable
> or use the name for some other purpose, that in itself is not illegal
> or fatal.
>
> However, nothing is created until you use -generate- later, directly
> or indirectly.
>
> A temporary name has various characteristics. It is referred to by a
> local macro with the name you supply, which holds a legal Stata name
> starting with two underscores and ending with 6 digits which are
> guaranteed not to be in use at the moment. When your program ends,
> Stata drops any such variables or scalars.
>
> As variables and scalars share the same name space, you could use
> -tempname- to the same end, and vice versa:
>
> . tempvar foo
>
> . scalar `foo' = 42
>
> . scalar list
> __000000 =         42
>
> . di "`foo'"
> __000000
>
> It is, however, likely to be considered poor style to use -tempvar-
> when you are creating scalars, and good style to use -tempvar- when
> creating variables.
>
> Nick
> [email protected]
>
>
> On 29 October 2013 23:56, David Rivenbark <[email protected]> wrote:
>> Hi all:
>>
>>         I have a simple user defined likelihood function that has the
>> following form
>>
>> Program my_lf
>>
>>         args lnfj theta1 theta2
>>         tempname xb sigma
>>         tempvar tmp1 tmp2 ....
>>
>>         gen double `xd' = `theta1'
>>         gen double `sigma' = exp(`theta2')
>>
>>         ...
>>
>>         replace `lnfj' = ...
>>
>> end
>>
>> My model also assumes that `theta1' has of two independent variables x1 and
>> x2.
>>
>>         I noticed that the results for my model are unchanged by using
>> -tempname- or -tempvar-. I assume this is because I specify "gen double `xb'
>> = `theta1'" Has anyone else noticed this before?
>>
>> David
>>
>>
>> *
>> *   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index