Please ! Stop using floating point numbers for financial data

Posted by Chris Burnley on July 3, 2009

Again and again I see developers using doubles for storing financial data (e.g. monetary figures, interest rates etc.). This always causes some sort of rounding issue down the line.

try this in your IDE:

double total = 0;
for (int i =0; i <  10; i++){
  total += 0.1;
}
System.out.println(total);

If you expect the answer to be 1 or 1.0, then you’re wrong !

Question: are you writing a game ? Answer: no; then use an effing BigDecimal!

Grrr…

Rate this:

Like this:

Be the first to like this post.

This entry was posted on July 3, 2009 at 10:55 am and is filed under Grrrr..., Java etc.... You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

2 Responses to “Please ! Stop using floating point numbers for financial data”

Jon Court said

July 3, 2009 at 7:42 pm
But, before you go creating BigDecimals from doubles all over the place be sure to read the javadoc here.

In summary – you would generally use:
```
BigDecimal.valueOf(double)
```
rather than:
```
new BigDecimal(double)
```
Reply
- Chris Burnley said
  
  July 3, 2009 at 8:00 pm
  The point is: don’t use them, ever if you can avoid it. The are only approximations. All input should be captured directly into bigdecimals.
  
  Reply