Please do not send HTML to the list!
Please do not send HTML to the list!
The problem is not, I think, one of
incompatibility.
My surmise is that -aflogit- is broken by
some recent tightening-up in Stata.
The problem is with references to
$`1'
where Stata is expected by the programmer
to take the local macro name first and then
evaluate the global name that results. If that
is what you want, you are expected to spell that
out explicitly.
Presumably this worked at one point, but
(again I surmise) StataCorp perhaps tightened
up on what was seen as a bug or a misfeature.
A quick experiment indicates that changes
to the code so that these lines read as below
produces output.
212 replace `1' = ${`1'} - `1'
241 gen `zmx' = ${`1'} - `1'
272 replace `1' = ${`1'} if `1' != .
281 replace `1' = ${`1'} if `1' != . /* reset exposure to reference level */
That is, assignments referring to
$`1'
should be to
${`1'}
instead.
Nick
[email protected]
-----Original Message-----
From: [email protected] [mailto:[email protected]]On Behalf Of rps
Sent: 11 August 2006 19:10
To: [email protected]
Subject: st: Variable names incompatable with aflogit?
I have encountered the following problem when using the aflogit (sbe21.pkg) user-supplied module for calculating population attributable fraction (AF) following a logistic regression (the data set is tuyns.dta, which is distributed with the package):
The following logistic regression command is issued:
xi: logit case i.alcohol [fw=n]
which produces:
i.alcohol _Ialcohol_0-3 (naturally coded; _Ialcohol_0 omitted)
Iteration 0: log likelihood = -494.74421
Iteration 1: log likelihood = -428.70187
Iteration 2: log likelihood = -421.84193
Iteration 3: log likelihood = -421.49571
Iteration 4: log likelihood = -421.49545
Logistic regression Number of obs = 975
LR chi2(3) = 146.50
Prob > chi2 = 0.0000
Log likelihood = -421.49545 Pseudo R2 = 0.1481
------------------------------------------------------------------------------
case | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_Ialcohol_1 | 1.27124 .232332 5.47 0.000 .8158777 1.726602
_Ialcohol_2 | 2.054459 .2611044 7.87 0.000 1.542704 2.566214
_Ialcohol_3 | 3.304162 .3236511 10.21 0.000 2.669817 3.938506
_cons | -2.588542 .1925445 -13.44 0.000 -2.965922 -2.211161
------------------------------------------------------------------------------
Then the aflogit command is given:
aflogit _Ialcohol* [fw=n], cc
which produces the following error:
Population attributable fraction from logistic regression
Case-control data (n=975)
Using weights [fweight = n]
Term Ref. A.F. s.e. [95% Conf. Int.]*
----------------------------------------------------------
l_1 not found
r(111);
Referring to the model terms explicitly does not help, as in
aflogit _Ialcohol_1 _Ialcohol_2 _Ialcohol_3 [fw=n], cc
However, renaming _Ialcohol_1 _Ialcohol_2 _Ialcohol_3 as alcohol1 alcohol2 alcohol3, respectively, allows the aflogit module to execute:
logit case alcohol1 alcohol2 alcohol3 [fw=n]
aflogit alcohol1 alcohol2 alcohol3 [fw=n], cc
producing
[logistic regression outout omitted]
Population attributable fraction from logistic regression
Case-control data (n=975)
Using weights [fweight = n]
Term Ref. A.F. s.e. [95% Conf. Int.]*
----------------------------------------------------------
alcohol1 0 0.2698 0.0451 0.1759 0.3530
alcohol2 0 0.2223 0.0340 0.1526 0.2863
alcohol3 0 0.2167 0.0287 0.1584 0.2710
----------------------------------------------------------
TOTAL 0.7089 0.0508 0.5903 0.7931
* CI calculated on log(1-AF) scale
Any insight anyone can give as to why aflogit doesn't accept the original indicator variable names will be appreciated. For the present example, renaming the variables is a reasonable solution, but it becomes tedious is there are several factors in the model each with multiple indicator variables.
Thanks,
Richard Steiner
The University of Akron
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
