Re: st: bug in -xtile-

[Nick Cox <njcoxstata@gmail.com>]

Now I agree with you. The code for -xtile- includes the test

		if `nquanti' > _N + 1 {
			di in red "nquantiles() must be less than or " /*
			*/ "equal to number of observations plus one"
			exit 198
		}

but a better test would be something like

marksample touse
qui count if `touse'

if `nquanti' > r(N) + 1 {

The total number of observations _N is not the right number to check against.

Nick

On Fri, Feb 1, 2013 at 3:21 AM, Dr. Yu Chen <profyuchen@gmail.com> wrote:

> Nick, if this is -xtile- point of view, then it will not work if I
> delete observations whose value<5000. See my code below.
> In other words, in the first example I added an if condition, and in
> the second example, I deleted those observations. Why is the result
> different? I think this is an inconsistency in -xtile-.
>
> sysuse auto,clear
> keep in 1/10
> drop if price<5000
> xtile decile=price , nq(10)

> On Thu, Jan 31, 2013 at 8:08 PM, Nick Cox <njcoxstata@gmail.com> wrote:
>> In your example 6 values qualify as < 5000. -xtile- classifies into 10
>> classes as best it can
>>
>> (Here -groups- is from SSC.)
>>
>> . groups price decile, sep(0)
>>
>>   +----------------------------------+
>>   | price   decile   Freq.   Percent |
>>   |----------------------------------|
>>   | 3,799        1       1     16.67 |
>>   | 4,082        2       1     16.67 |
>>   | 4,099        4       1     16.67 |
>>   | 4,453        6       1     16.67 |
>>   | 4,749        7       1     16.67 |
>>   | 4,816        9       1     16.67 |
>>   +----------------------------------+
>>
>> You seem to be saying that they should not be classified if there are
>> less than 10. That's a point of view, but it is not -xtile-'s point of
>> view.
>>
>> The values are of the new variable are not equally spaced integers,
>> but they could not be given the request.
>>
>> Nick
>>
>> On Fri, Feb 1, 2013 at 1:45 AM, Dr. Yu Chen <profyuchen@gmail.com> wrote:
>>> I think there is a bug in -xtile-. I ran the code below, and I still
>>> got deciles, but they are incomplete. I would expect missing values
>>> because there are not enough observations.
>>> I use Stata 12.
>>>
>>> sysuse auto,clear
>>> keep in 1/10
>>> xtile decile=price if price<5000, nq(10)
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/