The limitations of the -mod()- function
given fractional parts are indeed important.
FWIW, Tom Steichen and I wrote this up at
http://www.stata.com/support/faqs/data/mod.html
I was presupposing manipulations of integers
only, as the context implies.
Nick
[email protected]
Richard Atkins
> Rather than use the mod function I would recommend converting all your
> values to formatted strings using the =string(varname, format) command
> prior to extracting the digits.
>
> In general the mod function works well but I have come across datasets
> where values that displayed in tables and lists as (for example) 0.7
> produced a '6' when I tried to extract the 'tenths' digit using
> mod(int(varname*10),10). It turned out that the floating point
> representation of this value in the dta file was actually
> 0.69999999 so
> the mod function gave me exactly what I asked for rather than what I
> actually wanted.
>
> Some examples of extracting digits in this way are available in my
> digit preference functions at
> http://www.richardatkins.co.uk/statistics/statabits.htm
> >>> [email protected] 03/15/06 3:04 pm >>>
> -pnrcheck- from SSC covers the Swedish case.
>
> In this case, look _inside_ the program
> for some ideas. String functions are not
> the only way. -mod()- is also useful.
>
> The authors of -pnrcheck- are two Italians
> and a Welshman, just who you would expect
> to be working on Swedish identifiers.
>
> Nick
> [email protected]
>
> Lia Orfei
>
> > I have a variable that contains 7-digit numbers; the first
> > digit is the
> > remainder of the sum of the last six digits divided by 10
> > (check digit)
> >
> > example: 1030332 => 0+3+0+3+3+2=11/10=1+1 => check digit=1
> > 12050457 => 0+5+0+4+5+7=21/10=2+1 => check digit=1
> > and so on.
> >
> > Now I want to assess whther the check digit is correct for each
> entry.
> >
> > How do I calculate the partial sum of the contents of this variable?
> >
> > Extra challenge: some of the numbers sum up to 10, 20 etc so
> > the check
> > digit is actually 0 and it does not appear in the data for
> > some reason
> > (example 50456 is correct becasue 5+0+4+5+6=20 so the remainder is
> 0).
> >
> > Any ideas?? I was thinking about using the string functions.
*
* 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/
