as you can see,
assert float(.01)==.01
returns false so your
replace x_str = "0.01" if float(x) == 0.01
evaluates to false and there is no replacement
it may seem puzzling but try
. di %21x .01
+1.47ae147ae147bX-007
. di %21x float(.01)
+1.47ae140000000X-007
as you can see, the underlying hexadecimal representation of ..01 and float(.01) differs, see nick cox' stata tip 33 for details:
http://www.stata-journal.com/article.html?article=dm0022
and i also think bill gould has written about precision previously on stata list.
mike
--- On Wed, 6/17/09, David Airey <[email protected]> wrote:
> From: David Airey <[email protected]>
> Subject: st: float function question
> To: "Statalist" <[email protected]>
> Date: Wednesday, June 17, 2009, 2:55 AM
> .
>
> Why does the float function not work like this? I feel
> silly.
>
> clear
> input float x
> 0.00
> 0.01
> 0.03
> 0.10
> 0.30
> 1.00
> end
> generate str4 x_str = ""
> replace x_str = "0.00" if float(x) == 0.0
> replace x_str = "0.01" if float(x) == 0.01
> replace x_str = "0.03" if float(x) == 0.03
> replace x_str = "0.10" if float(x) == 0.1
> replace x_str = "0.30" if float(x) == 0.3
> replace x_str = "1.00" if float(x) == 1.0
> list, clean
>
> which gave
>
> . list, clean
>
>
> x x_str
> 1. 0 0.00
> 2. .01
> 3. .03
> 4. .1
> 5. .3
> 6. 1 1.00
>
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/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/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
