Index of /~taxsim/estate-tax/57-59
From wk2110@columbia.edu Sat Oct 7 11:11:05 2006
Date: Sat, 7 Oct 2006 11:11:03 -0400 (EDT)
From: Wojciech Kopczuk <wk2110 at columbia dot edu>
To: Daniel Feenberg <feenberg at nber dot org>
Subject: Re: estate tax data
This is IML error. Higher up in the log file you can find:
NOTE: End of file reached for FILENAME:MYDATA.
ERROR: Returning from module READN without return value.
According to SAS/IML documentation, when "input" encounters end-of-file in
the IML "do data" step (corresponding to the regular data step), the data
step ends. This is convenient, but in that particular case input is in a
user-defined function readn and end-of-file causes exit from both the
function and the data step immediately, without return value. One would have
to test for end-of-file in the function to fix it, but my IML knowledge is
very limited and I did not bother to find out if/how it can be done.
Wojciech
On Sat, 7 Oct 2006, Daniel Feenberg wrote:
>
>
> On Thu, 5 Oct 2006, Wojciech Kopczuk wrote:
>
> > The files are attached. The 57-9 file should work with just an
> > adjustment to the path (md5sum of data file on the nber server and my
> > machine is the same). The log file shows one error but I know why and it
> > has no consequences, so I
>
> I am curious - what causes the error? It claims to be on page 6, but there
> is no page 6.
>
> Dan
>
> > let it pass. The only thing that needed to be done in the program was to
> > change the byte order of numbers read (there may be a informat to deal
> > with it but I did not know what it is and I implemented a quick hack to
> > fix it). The program creates two files - estat579 and trust579, the
> > reason for that is that each person can have multiple trust records.
> >
> > The second program reads the 1977 file. What I have is the same as
> > your RG058.DEC.PUB7477.gz file (not the .txt file) so again only
> > adjustments to the path should be necessary.
> >
> > Feel free to use it as you wish.
> >
> > Best,
> >
> > Wojtek
> >
> >
> > On Thu, 5 Oct 2006, Daniel Feenberg wrote:
> >
> > >
> > > I'd be really interested in having the programs, and posting them on
> > > the web site, if that would be ok. Otherwise I can keep them private.
> > >
> > > Dan
> > >
> > > On Thu, 5 Oct 2006, Wojciech Kopczuk wrote:
> > >
> > > > Dear Dan,
> > > >
> > > > I've just come across your note on estate tax data from the National
> > > > Archives (http://www.nber.org/~taxsim/estate-tax/) If you recall, I
> > > > pointed you in the direction of these files a while ago and in fact
> > > > I've had both for almost three years. I was able to read them both,
> > > > including the 1959 that you seem to have had problems with. The 1977
> > > > one I used extensively, the paper the relies on it ("Bequest and tax
> > > > planning: evidence from estate tax returns,"
> > > > http://www.columbia.edu/~wk2110/bin/ill.pdf) will soon be posted as
> > > > nber working paper. I haven't really used the 1959 dataset for much.
> > > > Let me know if you want the programs or just feel free to add a note
> > > > to the data saying that I can help if anyone is interested
> > > >
> > > > Best,
> > > >
> > > > Wojciech
> > > >
> > > >
> > > >
> > > > --
> > > > Wojciech Kopczuk
> > > > Economics and SIPA
> > > > Columbia University
> > > > tel. 212-854-2519
> > > > www.kopczuk.com
> > > >
> > > >
> > >
> > >
> >
> > --
> > Wojciech Kopczuk
> > Economics and SIPA
> > Columbia University
> > tel. 212-854-2519
> > www.kopczuk.com
>
>
--
Wojciech Kopczuk
Economics and SIPA
Columbia University
tel. 212-854-2519
www.kopczuk.com