Truncate a number to two decimals

[Luciana Pando]

Hi, I have a doubt. I have a variable named "score" (float) as showed below and I want to truncate it to two decimals. I mean, I have values like 2.666667 and 2.1666667 and I want to generate a new variable that would be these numbers but truncated to two decimals. I want to get, for these examples, 2.66 and 2.16, not rounding. Thanks for your help.

   ID     score
"00002484"  2.833333
"00002484"         3
"00002484"         3
"00002484"         3
"00003861" 2.3333333
"00003861" 2.3333333
"00003861"      2.25
"00003861"      2.75
"00004010"  2.666667
"00004010"         3
"00004010"      2.75
"00004010"         3
"00005099" 2.1666667
"00005099"       2.5
"00005099"      1.75
"00007848" 2.3333333
"00007848" 2.3333333
"00007848"      1.75
"00007848"       2.5
"00015005" 1.8333334
"00015005"  2.833333
"00015005"      1.25
"00015005"         3
"00020453" 2.1666667

[Maarten Buis]

In principle the answer is gen score2 = floor(score*100)/100 For more see this Stata tip by Nick Cox: https://www.stata-journal.com/sjpdf....iclenum=dm0002

However, I assume you are aware of the pitfalls that come with precision, if not then this is a good starting point: https://blog.stata.com/2012/04/02/th...-to-precision/

[Nick Cox]

I agree with Maarten Buis. In practice

format score %3.2f

has all the advantages of showing or seeing two decimal places consistently with none of the disadvantages of arbitrarily mauling your data, even minutely.

[Maarten Buis]

Normally when people want the first two digits, they want it rounded, and in that case I agree with Nick Cox . However, Luciana asked for just the first two digits without rounding. In that case I don't know of a format that does that (nor do I think it would have enough applications for it to be worth while to implement).

[Marcos Almeida]

Just to play a little bit in Stata with some alternatives (clumsy as they might be in the particular case), this might well be another way to reach the goal:

tostring score, gen(myscore1) force
 
gen myscore2 = substr(myscore1, 1, 4)

destring myscore2, gen(myscore3)
 
list myscore* in 1/10

     +-----------------------------------+
     |    myscore1   myscore2   myscore3 |
     |-----------------------------------|
  1. | 2.833333015       2.83       2.83 |
  2. |           3          3          3 |
  3. |           3          3          3 |
  4. |           3          3          3 |
  5. | 2.333333254       2.33       2.33 |
     |-----------------------------------|
  6. | 2.333333254       2.33       2.33 |
  7. |        2.25       2.25       2.25 |
  8. |        2.75       2.75       2.75 |
  9. | 2.666666985       2.66       2.66 |
 10. |           3          3          3 |
     +-----------------------------------+

des

Contains data
  obs:            24                          
 vars:             5                          
 size:         1,128                          
----------------------------------------------------------------------------------------------------------
              storage   display    value
variable name   type    format     label      variable label
----------------------------------------------------------------------------------------------------------
ID              str20   %20s                  
score           float   %9.0g                
myscore1        str11   %11s                  score
myscore2        str4    %9s                  
myscore3        double  %10.0g                
----------------------------------------------------------------------------------------------------------
Sorted by:
     Note: Dataset has changed since last saved.

[Nick Cox]

Maarten Buis is right!

[Luciana Pando]

Thank you very much for your help! However, I was trying to use the command "floor" but Stata says "floor not found". Any idea of what could be happening?


Thanks a lot!


Luciana

[Nick Cox]

floor() is a function, not a command. In some software the distinction is not sharp, but in Stata it is.

More crucially for your question: We can't see any syntax in #7 but one or more spaces inserted between floor and the parenthesis ( would be problematic.

In this

. di floor (7)
floor not found
r(111);

. di floor(7.2)
7

the problem in the first command is that Stata can only guess that you are referring to a variable or scalar called floor -- and as reported it can't find one. Close up, and it is clear that you are referring to a function.

[William Lisowski]

In post #2 Maarten shows the floor() function used in an expression as part of the generate command (which he abbreviates to gen). There is no floor command.

Added in edit: crossed with Nick's post #8, in which he diagnoses the likely source of the error message.