Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Loop to replace values of a variable


From   Jeph Herrin <[email protected]>
To   [email protected]
Subject   Re: st: Loop to replace values of a variable
Date   Wed, 18 Dec 2013 14:00:27 -0500

Good point. So

 sort n
 replace x=x[1] + _n*.1

will do.


On 12/18/2013 11:22 AM, Nick Cox wrote:
In principle, yes. In practice, this means repeated addition of 0.1;
but 0.1 is a good example of a decimal that looks simple but can't be
held exactly as a binary. Do something like this

set obs 100
gen y1 = 0
replace y1 = y[_n-1] + 0.1 in 2/L
gen y2 = (_n-1)/10
edit y?

to convince yourself that the second method gives a better
approximation than the first.

More discussion at

SJ-10-1 pr0051  . . . . . . . . . . . . Stata tip 85: Looping over nonintegers
         . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  N. J. Cox
         Q1/10   SJ 10(1):160--163                                (no commands)
         tip on using forvalues to loop over numbers

http://www.stata-journal.com/sjpdf.html?articlenum=pr0051

Nick
[email protected]


On 18 December 2013 14:34, Jeph Herrin <[email protected]> wrote:
A solution that doesn't require knowing the value of x[1] in advance:

  sort x
  replace x=x[_n-1]+0.1 if _n>1

hth,
Jeph



On 12/18/2013 9:04 AM, emanuele mazzini wrote:

Yes, the way Marteen suggested worked greatly.
Thank you!!

2013/12/18 Fabrizio Colella - 1348010 <[email protected]>:

you generated x as Marteen suggested, now you just have to subtract:

gen x = _n*.1
gen y = -2.6+x

Fabrizio


----- Messaggio originale -----
Da: "Nick Cox" <[email protected]>
A: [email protected]
Sent: Mercoledì, 18 dicembre 2013 14:49:34
Oggetto: Re: st: Loop to replace values of a variable

For -2.5(0.1)2.1 or -2.5(0.1)2.5, you can still divide integers by 10.
No loop needed as _n always gives you integers in sequence.


Nick
[email protected]


On 18 December 2013 13:44, emanuele mazzini <[email protected]>
wrote:

Oh, yes, I see. Thank you very much for your answers, but I may have
oversimplified, since what you said would fit in this case, as x is
exactly = _n/10, but my problem is slightly more complex.

What if the values I want to replace go from -2.5 to 2.15, still with
a 0.1 interval? In this case I'd still need a loop, right?

2013/12/18 Maarten Buis <[email protected]>:

no need to loop:

gen x = _n*.1

Hope this helps,
Maarten


On Wed, Dec 18, 2013 at 2:33 PM, emanuele mazzini
<[email protected]> wrote:

Dear all statalist users,

I would like to program a (relatively) simple loop to replace the
values of a variable I created, but I cannot  figure it out how to
write the syntax.

The structure of my data is very simple:
I have 300 observations and a variable, say x, that in the first cell
(i.e. for n=1) is equal to 0.1, while it is missing for all other 299
cells. Data look like:

n     x
1    0.1
2     .
3     .
4     .
.
.
.
.
.
299  .
300  .

I would like the loop to replace the missing values of my variable by
going from cell 2 to cell 300, and to have a replace that would be at
a 0.1 intervals, so as to obtain data that look like:

n     x
1    0.1
2    0.2
3    0.3
4    0.4
.
.
.
.
.
299  29.9
300  30

I can obviously do it by hand, but a loop would be naturally much more
efficient!
Does anyone know how to make it?

*
*   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/


*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index