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: RE: assigning same number to slightly different string var.


From   Nick Cox <[email protected]>
To   "'[email protected]'" <[email protected]>
Subject   RE: st: RE: assigning same number to slightly different string var.
Date   Tue, 1 Nov 2011 13:26:34 +0000

If you had specified use of Stata 8, I would (probably) have remembered that -index()- is the function in your case. 

More importantly, the art of using the function is clearly to specify a substring to catch all the possibilities you want and none of the possibilities you don't want. Nothing rules out a compound condition with 

if index(myvar, "FTSE") & index(myvar, "UK") 

or even 

if index(myvar, "FTSE") & !index(myvar, "UK")

Nick 
[email protected] 

Philip Hantschk

Thanks for the quick answer. i work with stata 8, so i had to use index( instead of strpos(. 
It works fine, but i have a little problem left. For example an index is called FTSE Europe and another FTSE Europe ex UK. These are 2 different Indices and I want to assign different numbers, but when i write
replace newvar = 1 if index(myvar, "FTSE Europe") it changes both.

As I write this message I think I have a solution. I first change all FTSE to 1, then FTSE Europe to 1, then FTSE Europe ex UK to 2, then FTSE Europe USD to 3 etc. From short to long.

What do you think of that idea? or is there an easier way?

Philip Hantschk 

> Von: Nick Cox <[email protected]>

> You want a new  variable, which could be constructed like this 
> 
> gen newvar = . 
> replace newvar = 1 if strpos(myvar, "500") 
> replace newvar = 2 if strpos(myvar, "Dax") 
> ... 
> 
> -if strpos(,)- is equivalent to -if strpos(,) > 0-. 
> 
> tab myvar if missing(newvar) 
> 
> will tell you what you missed. 
> 
> You will also want to define value labels and assign them. 
> 
> Nick 
> [email protected] 
> 
> Philip Hantschk
> 
> i have a set of data containing names of indices, like S&P 500, Dax 30
> etc. I want to assign numbers to the indices, so that S&P 500=1, Dax 30=2 etc.
> The problem is, that the in the observations, S&P 500 has different ways
> of writing, e.g. Standard and Poor's 500, S&P 500, S&P 500 Index, etc. 
> 
> My question is, is there a way to assign numbers automatically, like tell
> Stata to give every observation that contains "500" the "1".
> 

*
*   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/


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